You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by tq...@apache.org on 2022/04/23 05:54:01 UTC
[tvm-site] branch asf-site updated: deploying docs (apache/tvm@bce57586bd3e41ea3c38a157c126f1fea40a8313)
This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a commit to branch asf-site
in repository https://gitbox.apache.org/repos/asf/tvm-site.git
The following commit(s) were added to refs/heads/asf-site by this push:
new 992ee47c3 deploying docs (apache/tvm@bce57586bd3e41ea3c38a157c126f1fea40a8313)
992ee47c3 is described below
commit 992ee47c3aa99b0d5b5a36d67448ad4062b2775b
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Sat Apr 23 05:53:56 2022 +0000
deploying docs (apache/tvm@bce57586bd3e41ea3c38a157c126f1fea40a8313)
---
.../from_oneflow.ipynb | 162 ++++
.../from_oneflow.py | 177 ++++
docs/_images/sphx_glr_from_oneflow_thumb.png | Bin 0 -> 26786 bytes
.../how_to/compile_models/from_mxnet.rst.txt | 2 +-
.../how_to/compile_models/from_oneflow.rst.txt | 311 +++++++
.../how_to/compile_models/from_paddle.rst.txt | 2 +-
.../how_to/compile_models/from_pytorch.rst.txt | 2 +-
.../how_to/compile_models/from_tensorflow.rst.txt | 2 +-
docs/_sources/how_to/compile_models/index.rst.txt | 20 +
.../compile_models/sg_execution_times.rst.txt | 21 +-
.../deploy_models/deploy_model_on_android.rst.txt | 2 +-
.../deploy_object_detection_pytorch.rst.txt | 4 +-
.../deploy_models/deploy_prequantized.rst.txt | 6 +-
.../deploy_prequantized_tflite.rst.txt | 4 +-
.../how_to/deploy_models/deploy_quantized.rst.txt | 2 +-
.../deploy_models/deploy_ssd_gluoncv.rst.txt | 4 +-
.../deploy_models/sg_execution_times.rst.txt | 16 +-
.../extend_tvm/bring_your_own_datatypes.rst.txt | 4 +-
.../how_to/extend_tvm/sg_execution_times.rst.txt | 10 +-
.../how_to/extend_tvm/use_pass_instrument.rst.txt | 16 +-
.../optimize_operators/opt_conv_cuda.rst.txt | 2 +-
.../optimize_operators/opt_conv_tensorcore.rst.txt | 2 +-
.../how_to/optimize_operators/opt_gemm.rst.txt | 16 +-
.../optimize_operators/sg_execution_times.rst.txt | 8 +-
.../sg_execution_times.rst.txt | 16 +-
.../tune_conv2d_layer_cuda.rst.txt | 895 ++++-----------------
.../tune_network_cuda.rst.txt | 2 +-
.../tune_network_x86.rst.txt | 4 +-
.../tune_sparse_x86.rst.txt | 227 +-----
.../tune_with_autotvm/sg_execution_times.rst.txt | 12 +-
.../tune_with_autotvm/tune_conv2d_cuda.rst.txt | 34 +-
.../work_with_microtvm/micro_autotune.rst.txt | 16 +-
.../work_with_microtvm/sg_execution_times.rst.txt | 12 +-
.../work_with_relay/sg_execution_times.rst.txt | 6 +-
.../work_with_schedules/sg_execution_times.rst.txt | 18 +-
.../how_to/work_with_schedules/tensorize.rst.txt | 2 +-
.../tutorials/autotvm/sg_execution_times.rst.txt | 6 +-
.../frontend/deploy_classification.rst.txt | 2 +-
.../tutorials/frontend/deploy_detection.rst.txt | 2 +-
.../tutorials/frontend/sg_execution_times.rst.txt | 6 +-
.../tutorials/optimize/sg_execution_times.rst.txt | 6 +-
.../topic/vta/tutorials/sg_execution_times.rst.txt | 6 +-
.../tutorial/auto_scheduler_matmul_x86.rst.txt | 6 +-
docs/_sources/tutorial/autotvm_relay_x86.rst.txt | 55 +-
.../tutorial/cross_compilation_and_rpc.rst.txt | 2 +-
docs/_sources/tutorial/intro_topi.rst.txt | 2 +-
docs/_sources/tutorial/sg_execution_times.rst.txt | 26 +-
.../tutorial/tensor_expr_get_started.rst.txt | 49 +-
docs/commit_hash | 2 +-
docs/how_to/compile_models/from_coreml.html | 1 +
docs/how_to/compile_models/from_darknet.html | 1 +
docs/how_to/compile_models/from_keras.html | 1 +
docs/how_to/compile_models/from_mxnet.html | 3 +-
.../{from_pytorch.html => from_oneflow.html} | 282 +++++--
docs/how_to/compile_models/from_onnx.html | 1 +
docs/how_to/compile_models/from_paddle.html | 7 +-
docs/how_to/compile_models/from_pytorch.html | 38 +-
docs/how_to/compile_models/from_tensorflow.html | 3 +-
docs/how_to/compile_models/from_tflite.html | 1 +
docs/how_to/compile_models/index.html | 7 +
docs/how_to/compile_models/sg_execution_times.html | 21 +-
docs/how_to/deploy/index.html | 4 +-
.../deploy_models/deploy_model_on_android.html | 2 +-
.../deploy_object_detection_pytorch.html | 106 ++-
docs/how_to/deploy_models/deploy_prequantized.html | 17 +-
.../deploy_models/deploy_prequantized_tflite.html | 4 +-
docs/how_to/deploy_models/deploy_quantized.html | 2 +-
docs/how_to/deploy_models/deploy_ssd_gluoncv.html | 36 +-
docs/how_to/deploy_models/sg_execution_times.html | 16 +-
.../extend_tvm/bring_your_own_datatypes.html | 4 +-
docs/how_to/extend_tvm/sg_execution_times.html | 10 +-
docs/how_to/extend_tvm/use_pass_instrument.html | 16 +-
docs/how_to/optimize_operators/opt_conv_cuda.html | 2 +-
.../optimize_operators/opt_conv_tensorcore.html | 2 +-
docs/how_to/optimize_operators/opt_gemm.html | 16 +-
.../optimize_operators/sg_execution_times.html | 8 +-
.../sg_execution_times.html | 14 +-
.../tune_conv2d_layer_cuda.html | 895 ++++-----------------
.../tune_with_autoscheduler/tune_network_cuda.html | 2 +-
.../tune_with_autoscheduler/tune_network_x86.html | 4 +-
.../tune_with_autoscheduler/tune_sparse_x86.html | 227 +-----
.../tune_with_autotvm/sg_execution_times.html | 12 +-
.../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 34 +-
docs/how_to/work_with_microtvm/micro_autotune.html | 16 +-
.../work_with_microtvm/sg_execution_times.html | 12 +-
.../how_to/work_with_relay/sg_execution_times.html | 6 +-
.../work_with_schedules/sg_execution_times.html | 18 +-
docs/how_to/work_with_schedules/tensorize.html | 2 +-
docs/objects.inv | Bin 22090 -> 22165 bytes
docs/reference/api/python/auto_scheduler.html | 4 +-
.../api/typedoc/classes/bytestreamreader.html | 12 +-
.../api/typedoc/classes/cachedcallstack.html | 34 +-
docs/reference/api/typedoc/classes/dldatatype.html | 12 +-
docs/reference/api/typedoc/classes/dldevice.html | 10 +-
.../reference/api/typedoc/classes/environment.html | 12 +-
docs/reference/api/typedoc/classes/ffilibrary.html | 20 +-
.../api/typedoc/classes/graphexecutor.html | 16 +-
docs/reference/api/typedoc/classes/instance.html | 40 +-
docs/reference/api/typedoc/classes/memory.html | 34 +-
docs/reference/api/typedoc/classes/module.html | 10 +-
docs/reference/api/typedoc/classes/ndarray.html | 22 +-
.../api/typedoc/classes/packedfunccell.html | 6 +-
docs/reference/api/typedoc/classes/rpcserver.html | 14 +-
docs/reference/api/typedoc/classes/scalar.html | 6 +-
.../api/typedoc/classes/webgpucontext.html | 12 +-
docs/reference/api/typedoc/enums/argtypecode.html | 30 +-
.../api/typedoc/enums/aynccallbackcode.html | 4 +-
.../api/typedoc/enums/dldatatypecode.html | 8 +-
.../api/typedoc/enums/rpcserverstate.html | 12 +-
docs/reference/api/typedoc/enums/sizeof.html | 18 +-
docs/reference/api/typedoc/index.html | 112 +--
.../api/typedoc/interfaces/disposable.html | 2 +-
.../api/typedoc/interfaces/functioninfo.html | 6 +-
.../api/typedoc/interfaces/libraryprovider.html | 4 +-
docs/searchindex.js | 2 +-
.../vta/tutorials/autotvm/sg_execution_times.html | 6 +-
.../tutorials/frontend/deploy_classification.html | 2 +-
.../vta/tutorials/frontend/deploy_detection.html | 2 +-
.../vta/tutorials/frontend/sg_execution_times.html | 6 +-
.../vta/tutorials/optimize/sg_execution_times.html | 6 +-
docs/topic/vta/tutorials/sg_execution_times.html | 6 +-
docs/tutorial/auto_scheduler_matmul_x86.html | 5 +-
docs/tutorial/autotvm_relay_x86.html | 168 ++--
docs/tutorial/cross_compilation_and_rpc.html | 2 +-
docs/tutorial/intro_topi.html | 2 +-
docs/tutorial/sg_execution_times.html | 26 +-
docs/tutorial/tensor_expr_get_started.html | 45 +-
127 files changed, 2061 insertions(+), 2739 deletions(-)
diff --git a/docs/_downloads/2e7b51cb39c472626dd3f046d9b89966/from_oneflow.ipynb b/docs/_downloads/2e7b51cb39c472626dd3f046d9b89966/from_oneflow.ipynb
new file mode 100644
index 000000000..8e31a012c
--- /dev/null
+++ b/docs/_downloads/2e7b51cb39c472626dd3f046d9b89966/from_oneflow.ipynb
@@ -0,0 +1,162 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [],
+ "source": [
+ "%matplotlib inline"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "\nCompile OneFlow Models\n======================\n**Author**: `Xiaoyu Zhang <https://github.com/BBuf/>`_\n\nThis article is an introductory tutorial to deploy OneFlow models with Relay.\n\nFor us to begin with, OneFlow package should be installed.\n\nA quick solution is to install via pip\n\n.. code-block:: bash\n\n pip install flowvision==0.1.0\n python3 -m pip install -f https://release.oneflow.info oneflow==0.7.0+cpu\n\nor please refer to official site:\nhttps://github. [...]
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [],
+ "source": [
+ "import os, math\nfrom matplotlib import pyplot as plt\nimport numpy as np\nfrom PIL import Image\n\n# oneflow imports\nimport flowvision\nimport oneflow as flow\nimport oneflow.nn as nn\n\nimport tvm\nfrom tvm import relay\nfrom tvm.contrib.download import download_testdata"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Load a pretrained OneFlow model and save model\n----------------------------------------------\n\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [],
+ "source": [
+ "model_name = \"resnet18\"\nmodel = getattr(flowvision.models, model_name)(pretrained=True)\nmodel = model.eval()\n\nmodel_dir = \"resnet18_model\"\nif not os.path.exists(model_dir):\n flow.save(model.state_dict(), model_dir)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Load a test image\n-----------------\nClassic cat example!\n\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [],
+ "source": [
+ "from PIL import Image\n\nimg_url = \"https://github.com/dmlc/mxnet.js/blob/main/data/cat.png?raw=true\"\nimg_path = download_testdata(img_url, \"cat.png\", module=\"data\")\nimg = Image.open(img_path).resize((224, 224))\n\n# Preprocess the image and convert to tensor\nfrom flowvision import transforms\n\nmy_preprocess = transforms.Compose(\n [\n transforms.Resize(256),\n transforms.CenterCrop(224),\n transforms.ToTensor(),\n transforms.Normalize(me [...]
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Import the graph to Relay\n-------------------------\nConvert OneFlow graph to Relay graph. The input name can be arbitrary.\n\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [],
+ "source": [
+ "class Graph(flow.nn.Graph):\n def __init__(self, module):\n super().__init__()\n self.m = module\n\n def build(self, x):\n out = self.m(x)\n return out\n\n\ngraph = Graph(model)\n_ = graph._compile(flow.randn(1, 3, 224, 224))\n\nmod, params = relay.frontend.from_oneflow(graph, model_dir)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Relay Build\n-----------\nCompile the graph to llvm target with given input specification.\n\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [],
+ "source": [
+ "target = tvm.target.Target(\"llvm\", host=\"llvm\")\ndev = tvm.cpu(0)\nwith tvm.transform.PassContext(opt_level=3):\n lib = relay.build(mod, target=target, params=params)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Execute the portable graph on TVM\n---------------------------------\nNow we can try deploying the compiled model on target.\n\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [],
+ "source": [
+ "target = \"cuda\"\nwith tvm.transform.PassContext(opt_level=10):\n intrp = relay.build_module.create_executor(\"graph\", mod, tvm.cuda(0), target)\n\nprint(type(img))\nprint(img.shape)\ntvm_output = intrp.evaluate()(tvm.nd.array(img.astype(\"float32\")), **params)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Look up synset name\n-------------------\nLook up prediction top 1 index in 1000 class synset.\n\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [],
+ "source": [
+ "synset_url = \"\".join(\n [\n \"https://raw.githubusercontent.com/Cadene/\",\n \"pretrained-models.pytorch/master/data/\",\n \"imagenet_synsets.txt\",\n ]\n)\nsynset_name = \"imagenet_synsets.txt\"\nsynset_path = download_testdata(synset_url, synset_name, module=\"data\")\nwith open(synset_path) as f:\n synsets = f.readlines()\n\nsynsets = [x.strip() for x in synsets]\nsplits = [line.split(\" \") for line in synsets]\nkey_to_classname = {spl[0]: \" [...]
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.7.5"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/docs/_downloads/f7ae979fbe61064749ce0fb7a621eb4c/from_oneflow.py b/docs/_downloads/f7ae979fbe61064749ce0fb7a621eb4c/from_oneflow.py
new file mode 100644
index 000000000..f92f0b0f1
--- /dev/null
+++ b/docs/_downloads/f7ae979fbe61064749ce0fb7a621eb4c/from_oneflow.py
@@ -0,0 +1,177 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements. See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership. The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied. See the License for the
+# specific language governing permissions and limitations
+# under the License.
+"""
+Compile OneFlow Models
+======================
+**Author**: `Xiaoyu Zhang <https://github.com/BBuf/>`_
+
+This article is an introductory tutorial to deploy OneFlow models with Relay.
+
+For us to begin with, OneFlow package should be installed.
+
+A quick solution is to install via pip
+
+.. code-block:: bash
+
+ pip install flowvision==0.1.0
+ python3 -m pip install -f https://release.oneflow.info oneflow==0.7.0+cpu
+
+or please refer to official site:
+https://github.com/Oneflow-Inc/oneflow
+
+Currently, TVM supports OneFlow 0.7.0. Other versions may be unstable.
+"""
+import os, math
+from matplotlib import pyplot as plt
+import numpy as np
+from PIL import Image
+
+# oneflow imports
+import flowvision
+import oneflow as flow
+import oneflow.nn as nn
+
+import tvm
+from tvm import relay
+from tvm.contrib.download import download_testdata
+
+######################################################################
+# Load a pretrained OneFlow model and save model
+# ----------------------------------------------
+model_name = "resnet18"
+model = getattr(flowvision.models, model_name)(pretrained=True)
+model = model.eval()
+
+model_dir = "resnet18_model"
+if not os.path.exists(model_dir):
+ flow.save(model.state_dict(), model_dir)
+
+######################################################################
+# Load a test image
+# -----------------
+# Classic cat example!
+from PIL import Image
+
+img_url = "https://github.com/dmlc/mxnet.js/blob/main/data/cat.png?raw=true"
+img_path = download_testdata(img_url, "cat.png", module="data")
+img = Image.open(img_path).resize((224, 224))
+
+# Preprocess the image and convert to tensor
+from flowvision import transforms
+
+my_preprocess = transforms.Compose(
+ [
+ transforms.Resize(256),
+ transforms.CenterCrop(224),
+ transforms.ToTensor(),
+ transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
+ ]
+)
+img = my_preprocess(img)
+img = np.expand_dims(img.numpy(), 0)
+
+######################################################################
+# Import the graph to Relay
+# -------------------------
+# Convert OneFlow graph to Relay graph. The input name can be arbitrary.
+class Graph(flow.nn.Graph):
+ def __init__(self, module):
+ super().__init__()
+ self.m = module
+
+ def build(self, x):
+ out = self.m(x)
+ return out
+
+
+graph = Graph(model)
+_ = graph._compile(flow.randn(1, 3, 224, 224))
+
+mod, params = relay.frontend.from_oneflow(graph, model_dir)
+
+######################################################################
+# Relay Build
+# -----------
+# Compile the graph to llvm target with given input specification.
+target = tvm.target.Target("llvm", host="llvm")
+dev = tvm.cpu(0)
+with tvm.transform.PassContext(opt_level=3):
+ lib = relay.build(mod, target=target, params=params)
+
+######################################################################
+# Execute the portable graph on TVM
+# ---------------------------------
+# Now we can try deploying the compiled model on target.
+target = "cuda"
+with tvm.transform.PassContext(opt_level=10):
+ intrp = relay.build_module.create_executor("graph", mod, tvm.cuda(0), target)
+
+print(type(img))
+print(img.shape)
+tvm_output = intrp.evaluate()(tvm.nd.array(img.astype("float32")), **params)
+
+#####################################################################
+# Look up synset name
+# -------------------
+# Look up prediction top 1 index in 1000 class synset.
+synset_url = "".join(
+ [
+ "https://raw.githubusercontent.com/Cadene/",
+ "pretrained-models.pytorch/master/data/",
+ "imagenet_synsets.txt",
+ ]
+)
+synset_name = "imagenet_synsets.txt"
+synset_path = download_testdata(synset_url, synset_name, module="data")
+with open(synset_path) as f:
+ synsets = f.readlines()
+
+synsets = [x.strip() for x in synsets]
+splits = [line.split(" ") for line in synsets]
+key_to_classname = {spl[0]: " ".join(spl[1:]) for spl in splits}
+
+class_url = "".join(
+ [
+ "https://raw.githubusercontent.com/Cadene/",
+ "pretrained-models.pytorch/master/data/",
+ "imagenet_classes.txt",
+ ]
+)
+class_name = "imagenet_classes.txt"
+class_path = download_testdata(class_url, class_name, module="data")
+with open(class_path) as f:
+ class_id_to_key = f.readlines()
+
+class_id_to_key = [x.strip() for x in class_id_to_key]
+
+# Get top-1 result for TVM
+top1_tvm = np.argmax(tvm_output.numpy()[0])
+tvm_class_key = class_id_to_key[top1_tvm]
+
+# Convert input to OneFlow variable and get OneFlow result for comparison
+with flow.no_grad():
+ torch_img = flow.from_numpy(img)
+ output = model(torch_img)
+
+ # Get top-1 result for OneFlow
+ top_oneflow = np.argmax(output.numpy())
+ oneflow_class_key = class_id_to_key[top_oneflow]
+
+print("Relay top-1 id: {}, class name: {}".format(top1_tvm, key_to_classname[tvm_class_key]))
+print(
+ "OneFlow top-1 id: {}, class name: {}".format(top_oneflow, key_to_classname[oneflow_class_key])
+)
diff --git a/docs/_images/sphx_glr_from_oneflow_thumb.png b/docs/_images/sphx_glr_from_oneflow_thumb.png
new file mode 100644
index 000000000..233f8e605
Binary files /dev/null and b/docs/_images/sphx_glr_from_oneflow_thumb.png differ
diff --git a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
index 208e7882c..548202ccf 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -98,7 +98,7 @@ In this section, we download a pretrained imagenet model and classify an image.
.. code-block:: none
- Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip33034771-7792-4214-abe4-5ba2d8d9fbb2 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+ Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip67880911-6b49-4538-938a-4b8f125626ab from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
x (1, 3, 224, 224)
diff --git a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
new file mode 100644
index 000000000..f674d0e8f
--- /dev/null
+++ b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
@@ -0,0 +1,311 @@
+.. note::
+ :class: sphx-glr-download-link-note
+
+ Click :ref:`here <sphx_glr_download_how_to_compile_models_from_oneflow.py>` to download the full example code
+.. rst-class:: sphx-glr-example-title
+
+.. _sphx_glr_how_to_compile_models_from_oneflow.py:
+
+
+Compile OneFlow Models
+======================
+**Author**: `Xiaoyu Zhang <https://github.com/BBuf/>`_
+
+This article is an introductory tutorial to deploy OneFlow models with Relay.
+
+For us to begin with, OneFlow package should be installed.
+
+A quick solution is to install via pip
+
+.. code-block:: bash
+
+ pip install flowvision==0.1.0
+ python3 -m pip install -f https://release.oneflow.info oneflow==0.7.0+cpu
+
+or please refer to official site:
+https://github.com/Oneflow-Inc/oneflow
+
+Currently, TVM supports OneFlow 0.7.0. Other versions may be unstable.
+
+
+.. code-block:: default
+
+ import os, math
+ from matplotlib import pyplot as plt
+ import numpy as np
+ from PIL import Image
+
+ # oneflow imports
+ import flowvision
+ import oneflow as flow
+ import oneflow.nn as nn
+
+ import tvm
+ from tvm import relay
+ from tvm.contrib.download import download_testdata
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ Out:
+
+ .. code-block:: none
+
+ /usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional_pil.py:193: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
+ def resize(img, size, interpolation=Image.BILINEAR):
+ /usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional.py:65: DeprecationWarning: NEAREST is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.NEAREST or Dither.NONE instead.
+ Image.NEAREST: "nearest",
+ /usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional.py:66: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
+ Image.BILINEAR: "bilinear",
+ /usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional.py:67: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.
+ Image.BICUBIC: "bicubic",
+ /usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional.py:68: DeprecationWarning: BOX is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BOX instead.
+ Image.BOX: "box",
+ /usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional.py:69: DeprecationWarning: HAMMING is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.HAMMING instead.
+ Image.HAMMING: "hamming",
+ /usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional.py:70: DeprecationWarning: LANCZOS is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.LANCZOS instead.
+ Image.LANCZOS: "lanczos",
+ /usr/local/lib/python3.7/dist-packages/flowvision/data/auto_augment.py:28: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
+ _RANDOM_INTERPOLATION = (Image.BILINEAR, Image.BICUBIC)
+ /usr/local/lib/python3.7/dist-packages/flowvision/data/auto_augment.py:28: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.
+ _RANDOM_INTERPOLATION = (Image.BILINEAR, Image.BICUBIC)
+
+
+
+Load a pretrained OneFlow model and save model
+----------------------------------------------
+
+
+.. code-block:: default
+
+ model_name = "resnet18"
+ model = getattr(flowvision.models, model_name)(pretrained=True)
+ model = model.eval()
+
+ model_dir = "resnet18_model"
+ if not os.path.exists(model_dir):
+ flow.save(model.state_dict(), model_dir)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ Out:
+
+ .. code-block:: none
+
+ Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
+
0%| | 0.00/41.5M [00:00<?, ?B/s]
0%| | 16.0k/41.5M [00:00<08:23, 86.3kB/s]
0%| | 48.0k/41.5M [00:00<05:20, 135kB/s]
0%| | 104k/41.5M [00:00<03:28, 208kB/s]
0%| | 208k/41.5M [00:00<02:07, 339kB/s]
1%| | 288k/41.5M [00:00<01:38, 437kB/s]
1%| | 416k/41.5M [00:00<01:08, 633kB/s]
1%|1 | 528k/41.5M [00:01<00:57, 744kB/s]
2%|1 | 672k/41.5M [00:01<00:46, 911kB/s]
2%|1 | 792k/41.5M [00:01<00:44, 957kB/s]
2%|2 | 936k/41.5M [00:01<00:40, 1.06MB/s]
2%|2 | 1.03M/41.5M [00:01<00:40, 1.06MB/s]
3%|2 | 1.19M/41.5M [00:01<00:35, 1.19MB/s]
3%|3 | 1.31M/41.5M [00:01<00:35, 1.17MB/s]
4%|3 | 1.48M/41.5M [00:01<00:31, 1.31MB/s]
4%|3 | 1.61M/41.5M [00:01<00:32, 1.28MB/s]
5%|4 | 1.91M/41.5M [00:02<00:30, 1.38MB/s]
5%|5 | 2.23M/41.5M [00:02<00:24, 1.7
0MB/s]
6%|5 | 2.41M/41.5M [00:02<00:23, 1.72MB/s]
6%|6 | 2.58M/41.5M [00:02<00:27, 1.47MB/s]
7%|6 | 2.77M/41.5M [00:02<00:25, 1.57MB/s]
7%|7 | 2.92M/41.5M [00:02<00:26, 1.52MB/s]
8%|7 | 3.12M/41.5M [00:02<00:24, 1.66MB/s]
8%|7 | 3.29M/41.5M [00:03<00:24, 1.64MB/s]
9%|8 | 3.68M/41.5M [00:03<00:21, 1.81MB/s]
10%|9 | 4.09M/41.5M [00:03<00:20, 1.92MB/s]
11%|# | 4.52M/41.5M [00:03<00:19, 2.03MB/s]
12%|#1 | 4.96M/41.5M [00:03<00:16, 2.35MB/s]
13%|#2 | 5.23M/41.5M [00:03<00:17, 2.23MB/s]
13%|#3 | 5.45M/41.5M [00:04<00:17, 2.12MB/s]
14%|#4 | 5.92M/41.5M [00:04<00:14, 2.50MB/s]
15%|#4 | 6.21M/41.5M [00:04<00:14, 2.60MB/s]
16%|#5 | 6.47M/41.5M [00:04<00:16, 2.26MB/s]
16%|#6 | 6.74M/41.5M [00:04<00:15, 2.39MB/s]
17%|#6 | 6.98M/41.5M [00:04<00:15, 2.33MB/s]
18%|#7 |
7.30M/41.5M [00:04<00:14, 2.54MB/s]
18%|#8 | 7.55M/41.5M [00:04<00:14, 2.47MB/s]
19%|#8 | 7.88M/41.5M [00:05<00:13, 2.70MB/s]
20%|#9 | 8.14M/41.5M [00:05<00:13, 2.67MB/s]
20%|## | 8.46M/41.5M [00:05<00:12, 2.86MB/s]
21%|##1 | 8.74M/41.5M [00:05<00:12, 2.79MB/s]
22%|##1 | 9.09M/41.5M [00:05<00:11, 2.93MB/s]
23%|##2 | 9.37M/41.5M [00:05<00:12, 2.80MB/s]
23%|##3 | 9.73M/41.5M [00:05<00:10, 3.05MB/s]
24%|##4 | 10.0M/41.5M [00:05<00:11, 2.96MB/s]
25%|##5 | 10.4M/41.5M [00:05<00:10, 3.23MB/s]
26%|##5 | 10.7M/41.5M [00:06<00:10, 3.17MB/s]
27%|##6 | 11.1M/41.5M [00:06<00:09, 3.38MB/s]
28%|##7 | 11.5M/41.5M [00:06<00:09, 3.32MB/s]
29%|##8 | 11.9M/41.5M [00:06<00:08, 3.56MB/s]
29%|##9 | 12.2M/41.5M [00:06<00:08, 3.55MB/s]
31%|### | 12.7M/41.5M [00:06<00:08, 3.72MB/s]
31%|###1 | 13.0M/41.5M [00:06<00:08, 3.
66MB/s]
32%|###2 | 13.5M/41.5M [00:06<00:07, 3.85MB/s]
33%|###3 | 13.8M/41.5M [00:06<00:07, 3.75MB/s]
34%|###4 | 14.3M/41.5M [00:06<00:07, 4.00MB/s]
35%|###5 | 14.7M/41.5M [00:07<00:07, 3.88MB/s]
37%|###6 | 15.2M/41.5M [00:07<00:06, 4.22MB/s]
38%|###7 | 15.6M/41.5M [00:07<00:06, 4.14MB/s]
39%|###8 | 16.1M/41.5M [00:07<00:05, 4.51MB/s]
40%|###9 | 16.6M/41.5M [00:07<00:05, 4.38MB/s]
41%|####1 | 17.1M/41.5M [00:07<00:05, 4.69MB/s]
42%|####2 | 17.6M/41.5M [00:07<00:05, 4.58MB/s]
44%|####3 | 18.1M/41.5M [00:07<00:05, 4.80MB/s]
45%|####4 | 18.6M/41.5M [00:07<00:05, 4.58MB/s]
46%|####6 | 19.2M/41.5M [00:08<00:04, 4.97MB/s]
47%|####7 | 19.7M/41.5M [00:08<00:04, 4.96MB/s]
49%|####8 | 20.3M/41.5M [00:08<00:04, 5.38MB/s]
50%|##### | 20.8M/41.5M [00:08<00:04, 5.38MB/s]
52%|#####1 | 21.5M/41.5M [00:08<00:03, 5.81MB/s]
53%|#####3 |
22.1M/41.5M [00:08<00:03, 5.79MB/s]
55%|#####4 | 22.7M/41.5M [00:08<00:03, 6.08MB/s]
56%|#####6 | 23.3M/41.5M [00:08<00:03, 6.05MB/s]
58%|#####7 | 24.0M/41.5M [00:08<00:02, 6.29MB/s]
59%|#####9 | 24.6M/41.5M [00:09<00:02, 6.26MB/s]
61%|######1 | 25.3M/41.5M [00:09<00:02, 6.55MB/s]
63%|######2 | 26.0M/41.5M [00:09<00:02, 6.51MB/s]
64%|######4 | 26.8M/41.5M [00:09<00:02, 6.90MB/s]
66%|######6 | 27.4M/41.5M [00:09<00:02, 6.79MB/s]
68%|######8 | 28.2M/41.5M [00:09<00:01, 7.25MB/s]
70%|######9 | 28.9M/41.5M [00:09<00:01, 7.14MB/s]
72%|#######2 | 29.9M/41.5M [00:09<00:01, 8.09MB/s]
74%|#######3 | 30.7M/41.5M [00:09<00:01, 8.05MB/s]
76%|#######5 | 31.5M/41.5M [00:09<00:01, 8.18MB/s]
78%|#######7 | 32.3M/41.5M [00:10<00:01, 7.88MB/s]
80%|#######9 | 33.1M/41.5M [00:10<00:01, 7.88MB/s]
81%|########1 | 33.8M/41.5M [00:10<00:01, 7.53MB/s]
84%|########3 | 34.7M/41.5M [00:10<00:00, 7
.60MB/s]
86%|########5 | 35.5M/41.5M [00:10<00:00, 7.87MB/s]
87%|########7 | 36.3M/41.5M [00:10<00:00, 6.91MB/s]
91%|######### | 37.6M/41.5M [00:10<00:00, 8.30MB/s]
93%|#########2| 38.4M/41.5M [00:10<00:00, 8.34MB/s]
95%|#########4| 39.2M/41.5M [00:11<00:00, 6.81MB/s]
97%|#########7| 40.4M/41.5M [00:11<00:00, 8.06MB/s]
99%|#########9| 41.2M/41.5M [00:11<00:00, 8.05MB/s]
100%|##########| 41.5M/41.5M [00:11<00:00, 3.84MB/s]
+
+
+
+Load a test image
+-----------------
+Classic cat example!
+
+
+.. code-block:: default
+
+ from PIL import Image
+
+ img_url = "https://github.com/dmlc/mxnet.js/blob/main/data/cat.png?raw=true"
+ img_path = download_testdata(img_url, "cat.png", module="data")
+ img = Image.open(img_path).resize((224, 224))
+
+ # Preprocess the image and convert to tensor
+ from flowvision import transforms
+
+ my_preprocess = transforms.Compose(
+ [
+ transforms.Resize(256),
+ transforms.CenterCrop(224),
+ transforms.ToTensor(),
+ transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
+ ]
+ )
+ img = my_preprocess(img)
+ img = np.expand_dims(img.numpy(), 0)
+
+
+
+
+
+
+
+Import the graph to Relay
+-------------------------
+Convert OneFlow graph to Relay graph. The input name can be arbitrary.
+
+
+.. code-block:: default
+
+ class Graph(flow.nn.Graph):
+ def __init__(self, module):
+ super().__init__()
+ self.m = module
+
+ def build(self, x):
+ out = self.m(x)
+ return out
+
+
+ graph = Graph(model)
+ _ = graph._compile(flow.randn(1, 3, 224, 224))
+
+ mod, params = relay.frontend.from_oneflow(graph, model_dir)
+
+
+
+
+
+
+
+Relay Build
+-----------
+Compile the graph to llvm target with given input specification.
+
+
+.. code-block:: default
+
+ target = tvm.target.Target("llvm", host="llvm")
+ dev = tvm.cpu(0)
+ with tvm.transform.PassContext(opt_level=3):
+ lib = relay.build(mod, target=target, params=params)
+
+
+
+
+
+
+
+Execute the portable graph on TVM
+---------------------------------
+Now we can try deploying the compiled model on target.
+
+
+.. code-block:: default
+
+ target = "cuda"
+ with tvm.transform.PassContext(opt_level=10):
+ intrp = relay.build_module.create_executor("graph", mod, tvm.cuda(0), target)
+
+ print(type(img))
+ print(img.shape)
+ tvm_output = intrp.evaluate()(tvm.nd.array(img.astype("float32")), **params)
+
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ Out:
+
+ .. code-block:: none
+
+ <class 'numpy.ndarray'>
+ (1, 3, 224, 224)
+
+
+
+Look up synset name
+-------------------
+Look up prediction top 1 index in 1000 class synset.
+
+
+.. code-block:: default
+
+ synset_url = "".join(
+ [
+ "https://raw.githubusercontent.com/Cadene/",
+ "pretrained-models.pytorch/master/data/",
+ "imagenet_synsets.txt",
+ ]
+ )
+ synset_name = "imagenet_synsets.txt"
+ synset_path = download_testdata(synset_url, synset_name, module="data")
+ with open(synset_path) as f:
+ synsets = f.readlines()
+
+ synsets = [x.strip() for x in synsets]
+ splits = [line.split(" ") for line in synsets]
+ key_to_classname = {spl[0]: " ".join(spl[1:]) for spl in splits}
+
+ class_url = "".join(
+ [
+ "https://raw.githubusercontent.com/Cadene/",
+ "pretrained-models.pytorch/master/data/",
+ "imagenet_classes.txt",
+ ]
+ )
+ class_name = "imagenet_classes.txt"
+ class_path = download_testdata(class_url, class_name, module="data")
+ with open(class_path) as f:
+ class_id_to_key = f.readlines()
+
+ class_id_to_key = [x.strip() for x in class_id_to_key]
+
+ # Get top-1 result for TVM
+ top1_tvm = np.argmax(tvm_output.numpy()[0])
+ tvm_class_key = class_id_to_key[top1_tvm]
+
+ # Convert input to OneFlow variable and get OneFlow result for comparison
+ with flow.no_grad():
+ torch_img = flow.from_numpy(img)
+ output = model(torch_img)
+
+ # Get top-1 result for OneFlow
+ top_oneflow = np.argmax(output.numpy())
+ oneflow_class_key = class_id_to_key[top_oneflow]
+
+ print("Relay top-1 id: {}, class name: {}".format(top1_tvm, key_to_classname[tvm_class_key]))
+ print(
+ "OneFlow top-1 id: {}, class name: {}".format(top_oneflow, key_to_classname[oneflow_class_key])
+ )
+
+
+
+
+.. rst-class:: sphx-glr-script-out
+
+ Out:
+
+ .. code-block:: none
+
+ Relay top-1 id: 281, class name: tabby, tabby cat
+ OneFlow top-1 id: 281, class name: tabby, tabby cat
+
+
+
+
+.. _sphx_glr_download_how_to_compile_models_from_oneflow.py:
+
+
+.. only :: html
+
+ .. container:: sphx-glr-footer
+ :class: sphx-glr-footer-example
+
+
+
+ .. container:: sphx-glr-download
+
+ :download:`Download Python source code: from_oneflow.py <from_oneflow.py>`
+
+
+
+ .. container:: sphx-glr-download
+
+ :download:`Download Jupyter notebook: from_oneflow.ipynb <from_oneflow.ipynb>`
+
+
+.. only:: html
+
+ .. rst-class:: sphx-glr-signature
+
+ `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/docs/_sources/how_to/compile_models/from_paddle.rst.txt b/docs/_sources/how_to/compile_models/from_paddle.rst.txt
index 335f91075..02b9f91ff 100644
--- a/docs/_sources/how_to/compile_models/from_paddle.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_paddle.rst.txt
@@ -201,7 +201,7 @@ Look up prediction top 1 index in 1000 class synset.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 9.448 seconds)
+ **Total running time of the script:** ( 1 minutes 11.807 seconds)
.. _sphx_glr_download_how_to_compile_models_from_paddle.py:
diff --git a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
index 080e1e4de..58514fbf7 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -79,7 +79,7 @@ Load a pretrained PyTorch model
.. code-block:: none
Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
-
0%| | 0.00/44.7M [00:00<?, ?B/s]
43%|####3 | 19.3M/44.7M [00:00<00:00, 203MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 235MB/s]
+
0%| | 0.00/44.7M [00:00<?, ?B/s]
1%|1 | 512k/44.7M [00:00<00:09, 4.97MB/s]
5%|4 | 2.04M/44.7M [00:00<00:03, 11.4MB/s]
9%|8 | 3.82M/44.7M [00:00<00:02, 14.7MB/s]
13%|#2 | 5.67M/44.7M [00:00<00:02, 16.5MB/s]
16%|#6 | 7.26M/44.7M [00:00<00:02, 14.0MB/s]
19%|#9 | 8.65M/44.7M [00:00<00:03, 11.9MB/s]
22%|##2 | 9.86M/44.7M [00:00<00:03, 10.7MB/s]
24%|##4 | 10.9M/44.7M [00:01<00:03, 10.1MB/s]
27%|##6 | 11.9M/44.7M [00:01<00:03, 10.0MB/s]
29%|##8 | 12.9M/44.7M [00:01<00:03, 9.61MB/s]
32%|###1 | 14.2M/44.7M [00:01<00:02, 10.7MB/s]
36%|###6 | 16.1M/44.7M [00:01<00:02, 13.1MB/s]
40%|#### | 17.9M/44.7M [00:01<00:02, 13.8MB/s]
43%|####3 | 19.2M/44.7M [00:01<00:02, 12.8MB/s]
46%|####6 | 20.7M/44.7M [00:01<00:01, 13.4MB/s]
49%|####9 | 22.0M/44.7M [00:01<00:01, 13.4MB/s]
52%|#####2 | 23.3M/44.7M [00:
01<00:01, 13.3MB/s]
55%|#####5 | 24.6M/44.7M [00:02<00:01, 12.1MB/s]
58%|#####7 | 25.8M/44.7M [00:02<00:01, 10.6MB/s]
60%|###### | 26.9M/44.7M [00:02<00:01, 11.0MB/s]
63%|######3 | 28.3M/44.7M [00:02<00:01, 11.8MB/s]
66%|######6 | 29.5M/44.7M [00:02<00:01, 10.8MB/s]
68%|######8 | 30.5M/44.7M [00:02<00:01, 8.73MB/s]
71%|####### | 31.6M/44.7M [00:02<00:01, 9.15MB/s]
73%|#######2 | 32.5M/44.7M [00:03<00:01, 9.05MB/s]
76%|#######5 | 33.8M/44.7M [00:03<00:01, 9.97MB/s]
78%|#######8 | 35.0M/44.7M [00:03<00:00, 10.4MB/s]
81%|######## | 36.0M/44.7M [00:03<00:00, 10.1MB/s]
84%|########3 | 37.5M/44.7M [00:03<00:00, 11.3MB/s]
88%|########8 | 39.3M/44.7M [00:03<00:00, 12.1MB/s]
91%|######### | 40.5M/44.7M [00:03<00:00, 10.3MB/s]
93%|#########3| 41.6M/44.7M [00:03<00:00, 10.5MB/s]
96%|#########5| 42.7M/44.7M [00:04<00:00, 9.42MB/s]
99%|#########8| 44.1M/44.7M [00:04<00:00, 9.00MB/s]
100%
|##########| 44.7M/44.7M [00:04<00:00, 11.0MB/s]
diff --git a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
index 0bf125797..914fc5cba 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -372,7 +372,7 @@ Run the corresponding model on tensorflow
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 1.729 seconds)
+ **Total running time of the script:** ( 1 minutes 3.238 seconds)
.. _sphx_glr_download_how_to_compile_models_from_tensorflow.py:
diff --git a/docs/_sources/how_to/compile_models/index.rst.txt b/docs/_sources/how_to/compile_models/index.rst.txt
index f6f14c0e2..538c13783 100644
--- a/docs/_sources/how_to/compile_models/index.rst.txt
+++ b/docs/_sources/how_to/compile_models/index.rst.txt
@@ -193,6 +193,26 @@ formats. These how-tos demostrate how to import models using the Python API.
:hidden:
/how_to/compile_models/from_paddle
+
+.. raw:: html
+
+ <div class="sphx-glr-thumbcontainer" tooltip="This article is an introductory tutorial to deploy OneFlow models with Relay.">
+
+.. only:: html
+
+ .. figure:: /how_to/compile_models/images/thumb/sphx_glr_from_oneflow_thumb.png
+
+ :ref:`sphx_glr_how_to_compile_models_from_oneflow.py`
+
+.. raw:: html
+
+ </div>
+
+
+.. toctree::
+ :hidden:
+
+ /how_to/compile_models/from_oneflow
.. raw:: html
<div style='clear:both'></div>
diff --git a/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt b/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
index 7b381665f..662e8dd4c 100644
--- a/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
@@ -5,14 +5,15 @@
Computation times
=================
-**04:54.129** total execution time for **how_to_compile_models** files:
+**05:32.910** total execution time for **how_to_compile_models** files:
-- **01:09.448**: :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)
-- **01:01.729**: :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``)
-- **00:57.601**: :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)
-- **00:25.458**: :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)
-- **00:22.648**: :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)
-- **00:21.624**: :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)
-- **00:19.162**: :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)
-- **00:13.923**: :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)
-- **00:02.536**: :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)
+- **01:11.807**: :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)
+- **01:03.238**: :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``)
+- **00:55.430**: :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)
+- **00:35.096**: :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``)
+- **00:25.403**: :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)
+- **00:22.619**: :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)
+- **00:21.671**: :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)
+- **00:21.390**: :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)
+- **00:13.578**: :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)
+- **00:02.679**: :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
index 44704bfea..e810c132e 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
@@ -393,7 +393,7 @@ Execute on TVM
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 16.0768 16.0997 16.1869 15.9239 0.0693
+ 15.5388 15.5206 15.6660 15.4596 0.0746
diff --git a/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt b/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
index fb54fd6f8..779756820 100644
--- a/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
@@ -108,7 +108,7 @@ Load pre-trained maskrcnn from torchvision and do tracing
.. code-block:: none
Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
-
0%| | 0.00/170M [00:00<?, ?B/s]
9%|9 | 15.5M/170M [00:00<00:00, 163MB/s]
22%|##2 | 37.7M/170M [00:00<00:00, 204MB/s]
36%|###6 | 61.6M/170M [00:00<00:00, 225MB/s]
50%|####9 | 84.7M/170M [00:00<00:00, 232MB/s]
64%|######3 | 108M/170M [00:00<00:00, 237MB/s]
77%|#######7 | 131M/170M [00:00<00:00, 239MB/s]
92%|#########2| 157M/170M [00:00<00:00, 249MB/s]
100%|##########| 170M/170M [00:00<00:00, 237MB/s]
+
0%| | 0.00/170M [00:00<?, ?B/s]
1%| | 0.99M/170M [00:00<00:17, 9.93MB/s]
1%|1 | 2.41M/170M [00:00<00:13, 12.7MB/s]
2%|2 | 3.63M/170M [00:00<00:16, 10.8MB/s]
3%|3 | 5.10M/170M [00:00<00:14, 12.2MB/s]
4%|3 | 6.29M/170M [00:00<00:15, 11.4MB/s]
5%|4 | 8.35M/170M [00:00<00:11, 14.4MB/s]
6%|6 | 10.6M/170M [00:00<00:09, 17.2MB/s]
7%|7 | 12.3M/170M [00:00<00:10, 16.0MB/s]
8%|8 | 13.9M/170M [00:01<00:11, 14.6MB/s]
9%|9 | 15.3M/170M [00:01<00:11, 14.0MB/s]
10%|9 | 16.7M/170M [00:01<00:13, 11.9MB/s]
11%|# | 18.4M/170M [00:01<00:12, 13.2MB/s]
13%|#3 | 22.3M/170M [00:01<00:07, 20.1MB/s]
14%|#4 | 24.4M/170M [00:01<00:07, 20.5MB/s]
16%|#5 | 26.4M/170M [00:01<00:08, 17.7MB/s]
17%|#6 | 28.3M/170M [00:01<00:08, 17.2MB/s]
18%|#7 | 30.1M/170M [00:02<00:09, 16.3MB/
s]
19%|#8 | 32.1M/170M [00:02<00:08, 17.4MB/s]
20%|#9 | 33.8M/170M [00:02<00:09, 14.7MB/s]
21%|## | 35.3M/170M [00:02<00:11, 12.8MB/s]
22%|##1 | 36.6M/170M [00:02<00:11, 12.5MB/s]
22%|##2 | 37.9M/170M [00:02<00:11, 11.7MB/s]
23%|##3 | 39.5M/170M [00:02<00:10, 12.9MB/s]
24%|##4 | 41.3M/170M [00:02<00:09, 14.3MB/s]
25%|##5 | 42.8M/170M [00:03<00:10, 12.7MB/s]
26%|##6 | 44.5M/170M [00:03<00:09, 14.0MB/s]
27%|##7 | 45.9M/170M [00:03<00:09, 13.6MB/s]
28%|##8 | 47.7M/170M [00:03<00:08, 14.9MB/s]
29%|##9 | 49.5M/170M [00:03<00:07, 15.8MB/s]
30%|### | 51.1M/170M [00:03<00:08, 15.0MB/s]
31%|### | 52.5M/170M [00:03<00:08, 15.1MB/s]
32%|###1 | 54.1M/170M [00:03<00:07, 15.3MB/s]
33%|###2 | 55.6M/170M [00:03<00:07, 15.6MB/s]
34%|###3 | 57.2M/170M [00:04<00:07, 15.1MB/s]
35%|###4 | 59.1M/170M [00:04<00:
07, 16.3MB/s]
36%|###6 | 61.2M/170M [00:04<00:06, 17.6MB/s]
37%|###7 | 62.9M/170M [00:04<00:06, 16.0MB/s]
38%|###7 | 64.5M/170M [00:04<00:07, 14.5MB/s]
39%|###9 | 67.0M/170M [00:04<00:06, 17.5MB/s]
41%|#### | 69.0M/170M [00:04<00:05, 18.2MB/s]
42%|####2 | 71.6M/170M [00:04<00:04, 20.7MB/s]
43%|####3 | 73.7M/170M [00:05<00:05, 19.7MB/s]
45%|####4 | 75.6M/170M [00:05<00:05, 16.8MB/s]
46%|####6 | 78.2M/170M [00:05<00:04, 19.5MB/s]
47%|####7 | 80.2M/170M [00:05<00:04, 18.9MB/s]
48%|####8 | 82.1M/170M [00:05<00:05, 18.1MB/s]
49%|####9 | 83.9M/170M [00:05<00:05, 18.0MB/s]
50%|##### | 85.7M/170M [00:05<00:05, 15.7MB/s]
51%|#####1 | 87.2M/170M [00:05<00:05, 15.1MB/s]
52%|#####2 | 88.7M/170M [00:06<00:06, 13.6MB/s]
53%|#####3 | 90.1M/170M [00:06<00:06, 13.6MB/s]
54%|#####3 | 91.7M/170M [00:06<00:05, 14.3MB/s]
55%|#####4 | 93.1M/170M
[00:06<00:05, 13.6MB/s]
56%|#####5 | 94.8M/170M [00:06<00:05, 14.6MB/s]
57%|#####6 | 96.5M/170M [00:06<00:04, 15.4MB/s]
58%|#####7 | 98.3M/170M [00:06<00:04, 16.2MB/s]
59%|#####8 | 99.9M/170M [00:06<00:05, 14.2MB/s]
60%|#####9 | 102M/170M [00:06<00:04, 15.7MB/s]
61%|###### | 103M/170M [00:07<00:05, 12.7MB/s]
62%|######2 | 105M/170M [00:07<00:04, 14.4MB/s]
63%|######2 | 107M/170M [00:07<00:05, 12.5MB/s]
64%|######4 | 109M/170M [00:07<00:04, 14.3MB/s]
65%|######4 | 110M/170M [00:07<00:04, 14.4MB/s]
66%|######6 | 112M/170M [00:07<00:03, 15.8MB/s]
67%|######7 | 114M/170M [00:07<00:03, 16.0MB/s]
68%|######8 | 116M/170M [00:07<00:03, 16.5MB/s]
69%|######9 | 117M/170M [00:08<00:03, 15.3MB/s]
70%|######9 | 119M/170M [00:08<00:03, 15.6MB/s]
71%|####### | 120M/170M [00:08<00:03, 15.6MB/s]
72%|#######1 | 122M/170M [00:08<00:03, 15.9MB/s]
73%|#######2 | 124M/170M [
00:08<00:03, 16.1MB/s]
74%|#######4 | 126M/170M [00:08<00:02, 17.2MB/s]
75%|#######5 | 127M/170M [00:08<00:02, 15.0MB/s]
76%|#######5 | 129M/170M [00:08<00:02, 15.2MB/s]
77%|#######6 | 130M/170M [00:09<00:03, 11.3MB/s]
78%|#######7 | 132M/170M [00:09<00:03, 12.5MB/s]
79%|#######8 | 134M/170M [00:09<00:02, 14.5MB/s]
80%|######## | 136M/170M [00:09<00:02, 16.0MB/s]
81%|########1 | 138M/170M [00:09<00:02, 15.5MB/s]
82%|########2 | 140M/170M [00:09<00:01, 16.9MB/s]
83%|########3 | 142M/170M [00:09<00:01, 15.6MB/s]
84%|########4 | 143M/170M [00:09<00:01, 15.6MB/s]
85%|########5 | 145M/170M [00:10<00:02, 13.0MB/s]
87%|########6 | 147M/170M [00:10<00:01, 15.8MB/s]
88%|########7 | 149M/170M [00:10<00:01, 17.1MB/s]
89%|########8 | 151M/170M [00:10<00:01, 13.0MB/s]
90%|########9 | 152M/170M [00:10<00:01, 14.2MB/s]
91%|######### | 154M/170M [00:10<00:01, 15.2MB/s]
92%|#########1| 156M/170M [00:10<0
0:01, 13.7MB/s]
93%|#########2| 158M/170M [00:10<00:00, 14.7MB/s]
94%|#########3| 159M/170M [00:11<00:00, 13.5MB/s]
95%|#########5| 162M/170M [00:11<00:00, 16.3MB/s]
97%|#########6| 164M/170M [00:11<00:00, 19.4MB/s]
98%|#########8| 167M/170M [00:11<00:00, 21.5MB/s]
100%|#########9| 169M/170M [00:11<00:00, 20.4MB/s]
100%|##########| 170M/170M [00:11<00:00, 15.4MB/s]
/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py:3878: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
for i in range(dim)
/usr/local/lib/python3.7/dist-packages/torchvision/models/detection/anchor_utils.py:127: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
@@ -253,7 +253,7 @@ Get boxes with score larger than 0.9
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 3 minutes 8.181 seconds)
+ **Total running time of the script:** ( 3 minutes 9.500 seconds)
.. _sphx_glr_download_how_to_deploy_models_deploy_object_detection_pytorch.py:
diff --git a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
index bc66c845b..5621f9b4d 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -187,7 +187,7 @@ training. Other models require a full post training calibration.
.. code-block:: none
Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
-
0%| | 0.00/13.6M [00:00<?, ?B/s]
23%|##2 | 3.05M/13.6M [00:00<00:00, 32.0MB/s]
45%|####5 | 6.11M/13.6M [00:00<00:00, 27.3MB/s]
100%|##########| 13.6M/13.6M [00:00<00:00, 54.2MB/s]
+
0%| | 0.00/13.6M [00:00<?, ?B/s]
7%|7 | 1.01M/13.6M [00:00<00:01, 10.3MB/s]
16%|#6 | 2.19M/13.6M [00:00<00:01, 11.4MB/s]
28%|##8 | 3.85M/13.6M [00:00<00:00, 14.1MB/s]
38%|###8 | 5.20M/13.6M [00:00<00:00, 10.7MB/s]
50%|####9 | 6.75M/13.6M [00:00<00:00, 11.7MB/s]
59%|#####8 | 7.94M/13.6M [00:00<00:00, 11.3MB/s]
71%|####### | 9.59M/13.6M [00:00<00:00, 12.9MB/s]
80%|######## | 10.9M/13.6M [00:00<00:00, 11.8MB/s]
89%|########9 | 12.1M/13.6M [00:01<00:00, 12.0MB/s]
100%|##########| 13.6M/13.6M [00:01<00:00, 12.2MB/s]
@@ -344,7 +344,7 @@ Here we give an example of how to measure performance of TVM compiled models.
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 88.7699 88.7508 92.6485 87.9840 0.6470
+ 90.0302 89.9670 91.2822 89.8079 0.2209
@@ -384,7 +384,7 @@ TODO
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 5.984 seconds)
+ **Total running time of the script:** ( 1 minutes 4.515 seconds)
.. _sphx_glr_download_how_to_deploy_models_deploy_prequantized.py:
diff --git a/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt b/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
index 6491d67c6..96dd99671 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
@@ -351,7 +351,7 @@ Here we give an example of how to measure performance of TVM compiled models.
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 121.0073 120.9071 122.0060 120.2682 0.4645
+ 118.0240 117.8995 122.2729 116.1341 0.9010
@@ -385,7 +385,7 @@ Here we give an example of how to measure performance of TVM compiled models.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 52.710 seconds)
+ **Total running time of the script:** ( 1 minutes 52.119 seconds)
.. _sphx_glr_download_how_to_deploy_models_deploy_prequantized_tflite.py:
diff --git a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
index cc4bb7520..49b5032b5 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -221,7 +221,7 @@ We create a Relay VM to build and execute the model.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 9.701 seconds)
+ **Total running time of the script:** ( 1 minutes 36.736 seconds)
.. _sphx_glr_download_how_to_deploy_models_deploy_quantized.py:
diff --git a/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt b/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
index bf9fcbff6..39390777e 100644
--- a/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
@@ -137,7 +137,7 @@ Convert and compile model for CPU.
data: None
input_sym_arg_type = in_param.infer_type()[0]
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
-
0%| | 0/132723 [00:00<?, ?KB/s]
1%|1 | 1479/132723 [00:00<00:09, 14378.80KB/s]
4%|3 | 4950/132723 [00:00<00:04, 25952.63KB/s]
8%|8 | 10996/132723 [00:00<00:02, 41525.98KB/s]
14%|#4 | 18751/132723 [00:00<00:02, 55645.45KB/s]
21%|## | 27503/132723 [00:00<00:01, 67083.48KB/s]
27%|##7 | 36367/132723 [00:00<00:01, 74388.69KB/s]
34%|###4 | 45201/132723 [00:00<00:01, 78937.14KB/s]
41%|#### | 54041/132723 [00:00<00:00, 81938.85KB/s]
47%|####7 | 63017/132723 [00:00<00:00, 84376.82KB/s]
54%|#####4 | 71905/132723 [00:01<00:00, 85764.19KB/s]
61%|###### | 80832/132723 [00:01<00:00, 86833.18KB/s]
68%|######7 | 89654/132723 [00:01<00:00, 87252.39KB/s]
74%|#######4 | 98382/132723 [00:01<00:00, 82998.12KB/s]
80%|######## | 106743/132723 [00:01<00:00, 83173.92KB/s]
87%|########7 | 115536/132723 [00:01<00:00, 84570.54KB/s]
94%|#########3
| 124404/132723 [00:01<00:00, 85782.74KB/s]
100%|##########| 132723/132723 [00:01<00:00, 77340.97KB/s]
+
0%| | 0/132723 [00:00<?, ?KB/s]
5%|4 | 6368/132723 [00:00<00:01, 63672.32KB/s]
11%|#1 | 14897/132723 [00:00<00:01, 76385.98KB/s]
18%|#7 | 23479/132723 [00:00<00:01, 80689.22KB/s]
24%|##4 | 32087/132723 [00:00<00:01, 82814.23KB/s]
31%|### | 40770/132723 [00:00<00:01, 84259.53KB/s]
37%|###7 | 49384/132723 [00:00<00:00, 84894.82KB/s]
44%|####3 | 57945/132723 [00:00<00:00, 85125.32KB/s]
50%|##### | 66458/132723 [00:00<00:00, 71183.84KB/s]
56%|#####5 | 73939/132723 [00:01<00:00, 60487.81KB/s]
61%|###### | 80456/132723 [00:01<00:00, 60996.03KB/s]
67%|######6 | 88391/132723 [00:01<00:00, 65757.99KB/s]
73%|#######3 | 96963/132723 [00:01<00:00, 71150.09KB/s]
80%|#######9 | 105644/132723 [00:01<00:00, 75512.47KB/s]
86%|########6 | 114291/132723 [00:01<00:00, 78627.39KB/s]
93%|#########2| 122937/132723 [00:01<00:00, 80888.66KB/s]
99%|########
#9| 131597/132723 [00:01<00:00, 82556.80KB/s]
100%|##########| 132723/132723 [00:01<00:00, 75800.40KB/s]
@@ -202,7 +202,7 @@ Display result
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 24.945 seconds)
+ **Total running time of the script:** ( 2 minutes 19.867 seconds)
.. _sphx_glr_download_how_to_deploy_models_deploy_ssd_gluoncv.py:
diff --git a/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt b/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
index 646a538b6..d48b95680 100644
--- a/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
@@ -5,13 +5,13 @@
Computation times
=================
-**10:31.937** total execution time for **how_to_deploy_models** files:
+**10:52.604** total execution time for **how_to_deploy_models** files:
-- **03:08.181**: :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``)
-- **02:24.945**: :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)
-- **01:52.710**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)
-- **01:09.701**: :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)
-- **01:05.984**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)
-- **00:28.405**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)
-- **00:21.821**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)
+- **03:09.500**: :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``)
+- **02:19.867**: :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)
+- **01:52.119**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)
+- **01:36.736**: :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)
+- **01:04.515**: :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)
+- **00:28.540**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)
+- **00:21.137**: :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)
- **00:00.190**: :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``)
diff --git a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
index 9ee8ce9e9..f48dd01b2 100644
--- a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
@@ -423,7 +423,7 @@ First let us define two helper functions to get the mobilenet model and a cat im
.. code-block:: none
- Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zipd92d2127-e27d-4c49-b468-f6776fd76cd2 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+ Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zipebb22811-12a9-49a8-a8d7-800ae10725cb from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
@@ -525,7 +525,7 @@ Now, to actually convert the entire network, we have written `a pass in Relay <h
.. code-block:: none
- Check failed: (lower) is false: FloatImm lowering function for target llvm type 150 not found
+ Check failed: (lower) is false: Intrinsic lowering function for target llvm, intrinsic name tir.sqrt, type 150 not found
diff --git a/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt b/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
index 988a65ab7..08f61fd83 100644
--- a/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
@@ -5,9 +5,9 @@
Computation times
=================
-**00:38.042** total execution time for **how_to_extend_tvm** files:
+**00:39.769** total execution time for **how_to_extend_tvm** files:
-- **00:34.591**: :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``)
-- **00:02.217**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)
-- **00:01.032**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)
-- **00:00.202**: :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``)
+- **00:36.149**: :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``)
+- **00:02.336**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)
+- **00:01.084**: :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)
+- **00:00.200**: :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``)
diff --git a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
index 7739248bc..d4fb58b2e 100644
--- a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
@@ -199,10 +199,10 @@ profile the execution time of each passes.
.. code-block:: none
Printing results of timing profile...
- InferType: 5919us [5919us] (45.59%; 45.59%)
- FoldScaleAxis: 7065us [2us] (54.41%; 54.41%)
- FoldConstant: 7063us [1481us] (54.40%; 99.97%)
- InferType: 5581us [5581us] (42.99%; 79.03%)
+ InferType: 6301us [6301us] (45.44%; 45.44%)
+ FoldScaleAxis: 7566us [2us] (54.56%; 54.56%)
+ FoldConstant: 7564us [1556us] (54.55%; 99.97%)
+ InferType: 6008us [6008us] (43.33%; 79.43%)
@@ -239,10 +239,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
.. code-block:: none
Printing results of timing profile...
- InferType: 5642us [5642us] (44.48%; 44.48%)
- FoldScaleAxis: 7044us [2us] (55.52%; 55.52%)
- FoldConstant: 7042us [1448us] (55.51%; 99.97%)
- InferType: 5594us [5594us] (44.09%; 79.44%)
+ InferType: 6063us [6063us] (44.71%; 44.71%)
+ FoldScaleAxis: 7497us [2us] (55.29%; 55.29%)
+ FoldConstant: 7495us [1531us] (55.27%; 99.97%)
+ InferType: 5964us [5964us] (43.98%; 79.57%)
diff --git a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
index 22b3c9443..48ad89a57 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
@@ -295,7 +295,7 @@ latency of convolution.
.. code-block:: none
- Convolution: 54.266551 ms
+ Convolution: 47.082751 ms
diff --git a/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt b/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
index 1b1b6010e..de8f97adf 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
@@ -628,7 +628,7 @@ be able to run on our build server
.. code-block:: none
- conv2d with tensor core: 7.173443 ms
+ conv2d with tensor core: 6.855349 ms
diff --git a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
index ba23a3b1c..029a23112 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -118,8 +118,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
.. code-block:: none
- Numpy running time: 0.019053
- Baseline: 3.195089
+ Numpy running time: 0.018223
+ Baseline: 3.524503
@@ -210,7 +210,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
.. code-block:: none
- Opt1: 0.298429
+ Opt1: 0.297080
@@ -309,7 +309,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
.. code-block:: none
- Opt2: 0.335815
+ Opt2: 0.328934
@@ -401,7 +401,7 @@ the access pattern for A matrix is more cache friendly.
.. code-block:: none
- Opt3: 0.116504
+ Opt3: 0.114862
@@ -520,7 +520,7 @@ flattening.
.. code-block:: none
- Opt4: 0.109173
+ Opt4: 0.111236
@@ -638,7 +638,7 @@ write to C when all the block results are ready.
.. code-block:: none
- Opt5: 0.109061
+ Opt5: 0.111643
@@ -759,7 +759,7 @@ Futhermore, we can also utilize multi-core processors to do the thread-level par
.. code-block:: none
- Opt6: 0.141408
+ Opt6: 0.144929
diff --git a/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt b/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
index a7917c236..e3854140f 100644
--- a/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
@@ -5,8 +5,8 @@
Computation times
=================
-**00:34.103** total execution time for **how_to_optimize_operators** files:
+**00:35.179** total execution time for **how_to_optimize_operators** files:
-- **00:31.489**: :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)
-- **00:01.395**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``)
-- **00:01.219**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)
+- **00:32.555**: :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)
+- **00:01.396**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``)
+- **00:01.228**: :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
index c05ada1db..7b25e3f06 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
@@ -5,11 +5,11 @@
Computation times
=================
-**05:14.315** total execution time for **how_to_tune_with_autoscheduler** files:
-
-- **02:39.840**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``)
-- **01:18.913**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)
-- **00:40.055**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)
-- **00:18.511**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)
-- **00:08.603**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)
-- **00:08.392**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)
+**05:02.667** total execution time for **how_to_tune_with_autoscheduler** files:
+
+- **02:27.183**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``)
+- **01:20.913**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)
+- **00:40.029**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)
+- **00:17.815**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)
+- **00:08.462**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)
+- **00:08.266**: :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
index ec11ae9e3..c73231e9b 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
@@ -187,7 +187,7 @@ file and apply it.
.. code-block:: none
- .T
+
@@ -222,12 +222,12 @@ cooperative fetching, unrolling and operator fusion.
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 128;
- allocate(conv2d_nchw: Pointer(local float32), float32, [28]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [1008]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [192]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [28], [], scope="local", align=64)[0] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 16;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [14]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [504]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [768]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [14], [], scope="local", align=32)[0] = 0f32
conv2d_nchw_1[7] = 0f32
conv2d_nchw_1[1] = 0f32
conv2d_nchw_1[8] = 0f32
@@ -241,476 +241,77 @@ cooperative fetching, unrolling and operator fusion.
conv2d_nchw_1[12] = 0f32
conv2d_nchw_1[6] = 0f32
conv2d_nchw_1[13] = 0f32
- conv2d_nchw_1[14] = 0f32
- conv2d_nchw_1[21] = 0f32
- conv2d_nchw_1[15] = 0f32
- conv2d_nchw_1[22] = 0f32
- conv2d_nchw_1[16] = 0f32
- conv2d_nchw_1[23] = 0f32
- conv2d_nchw_1[17] = 0f32
- conv2d_nchw_1[24] = 0f32
- conv2d_nchw_1[18] = 0f32
- conv2d_nchw_1[25] = 0f32
- conv2d_nchw_1[19] = 0f32
- conv2d_nchw_1[26] = 0f32
- conv2d_nchw_1[20] = 0f32
- conv2d_nchw_1[27] = 0f32
- for (rc.outer.outer: int32, 0, 32) {
+ for (rc.outer.outer: int32, 0, 64) {
for (rx.outer.outer: int32, 0, 3) {
- let cse_var_1: int32 = (rc.outer.outer*144)
+ let cse_var_2: int32 = (rc.outer.outer*392)
+ let cse_var_1: int32 = (rc.outer.outer*72)
{
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [1008], [], scope="shared")[threadIdx.x_1] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 7)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) - 1)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 14)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 6)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 21)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 13)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 28)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 20)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 35)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 27)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 42)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 34)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 49)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 41)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 56)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 63)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 70)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 48)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 77)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 55)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 84)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 62)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 91)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 69)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 76)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 105)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 83)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 112)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 90)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 119)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 126)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 133)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 97)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 140)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 104)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 147)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 111)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 154)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 118)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 161)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 125)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 168)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 132)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 175)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 139)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 182)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 189)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 146)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 203)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 153)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 210)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 160)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 217)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 167)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 224)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 174)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 231)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 181)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 238)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 188)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 245)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 252)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 259)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 195)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 266)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 202)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 273)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 209)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 280)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 216)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 287)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 223)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 294)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 230)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 301)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 237)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 308)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 315)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 322)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 244)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 329)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 251)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 336)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 258)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 343)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 265)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 350)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 272)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 357)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 279)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 364)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 286)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 371)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 378)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 385)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 293)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 300)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 399)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 307)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 406)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 314)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 413)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 321)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 420)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 328)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 427)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 335)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 434)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 441)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 448)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 342)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 455)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 349)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 462)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 356)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 469)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 363)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 476)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 370)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 483)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 377)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 490)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 384)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 497)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 504)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 511)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 391)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 518)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 398)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 525)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 405)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 532)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 412)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 539)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 419)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 546)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 426)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 553)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 433)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 560)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 567)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 574)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 440)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 581)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 447)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 588)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 454)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 595)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 461)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 602)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 468)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 609)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 475)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 616)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 482)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 623)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 630)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 637)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 489)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 644)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 496)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 651)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 503)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 658)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 510)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 665)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 517)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 672)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 524)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 679)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 531)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 686)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 693)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 700)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 538)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 707)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 545)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 714)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 552)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 721)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 559)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 728)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 566)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 735)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 573)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 742)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 580)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 749)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 756)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 763)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 587)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 770)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 594)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 777)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 601)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 784)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 608)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 791)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 615)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 798)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 622)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 805)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 629)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 812)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 819)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 826)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 636)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 833)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 643)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 840)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 650)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 847)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 657)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 854)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 664)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 861)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 671)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 868)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 678)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 875)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 882)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 889)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 685)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 896)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 692)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 903)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 699)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 910)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 706)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 917)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 713)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 924)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 720)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 931)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 727)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 938)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 945)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 952)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 734)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 959)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 741)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 966)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 748)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 973)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 755)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 980)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 762)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 987)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 769)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 994)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 776)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 1001)] = 0f32
- attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1: Buffer(kernel.shared, float32, [192], [], scope="shared")[threadIdx.x_2] = kernel[((((blockIdx.x*18432) + cse_var_1) + (threadIdx.x_2*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 7)] = kernel[((((blockIdx.x*18432) + cse_var_1) + ((threadIdx.x_2 + 7)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 14)] = kernel[((((blockIdx.x*18432) + cse_var_1) + ((threadIdx.x_2 + 14)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 21)] = kernel[((((blockIdx.x*18432) + cse_var_1) + ((threadIdx.x_2 + 21)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 28)] = kernel[((((blockIdx.x*18432) + cse_var_1) + ((threadIdx.x_2 + 28)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 35)] = kernel[((((blockIdx.x*18432) + cse_var_1) + ((threadIdx.x_2 + 35)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 42)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 42), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 42), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 49)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 49), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 1), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 56)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 56), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 8), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 63)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 63), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 15), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 70)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 70), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 22), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 77)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 77), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 29), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 84)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 84), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 36), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 91)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 91), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 43), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 98)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 98), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 2), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 105)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 105), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 9), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 112), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 16), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 119)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 119), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 23), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 126)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 126), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 30), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 133)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 133), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 37), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 140)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 140), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 44), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 147)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 147), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 3), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 154)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 154), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 10), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 161)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 161), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 17), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 168)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 168), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 24), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 175)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 175), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 31), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 182)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 182), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 38), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- if @tir.likely((threadIdx.x_2 < 3), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 189)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 189), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 45), 48)*3)) + rx.outer.outer)]
+ attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [504], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else(((((7 <= floormod(threadIdx.x_1, 63)) && (floormod(threadIdx.x_1, 63) < 56)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[((((cse_var_2 + (floordiv(threadIdx.x_1, 63)*49)) + rx.outer.outer) + floormod(threadIdx.x_1, 63)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ pad_temp.shared_1[(threadIdx.x_1 + 112)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((floordiv(threadIdx.x_1, 7) + 16), 9)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dt [...]
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ pad_temp.shared_1[(threadIdx.x_1 + 224)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((floordiv(threadIdx.x_1, 7) + 32), 9)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dt [...]
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ pad_temp.shared_1[(threadIdx.x_1 + 336)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 3), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((floordiv(threadIdx.x_1, 7) + 48), 9)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dt [...]
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ if @tir.likely((threadIdx.x_1 < 56), dtype=bool) {
+ pad_temp.shared_1[(threadIdx.x_1 + 448)] = @tir.if_then_else((((floormod((floordiv(threadIdx.x_1, 7) + 1), 9) < 8) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((floordiv(threadIdx.x_1, 7) + 64), 9)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 1), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+ }
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ kernel.shared_1: Buffer(kernel.shared, float32, [768], [], scope="shared")[threadIdx.x_2] = kernel[(((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_1) + (floormod(threadIdx.x_2, 24)*3)) + rx.outer.outer)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[(((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 8) + 14), 3)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 16), 24)*3)) + rx.outer.outer)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ kernel.shared_1[(threadIdx.x_2 + 224)] = kernel[(((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 8) + 28), 3)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 8), 24)*3)) + rx.outer.outer)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ kernel.shared_1[(threadIdx.x_2 + 336)] = kernel[((((((blockIdx.x*147456) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_1) + (floormod(threadIdx.x_2, 24)*3)) + rx.outer.outer) + 64512)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[(((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 8) + 56), 3)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 16), 24)*3)) + rx.outer.outer)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ kernel.shared_1[(threadIdx.x_2 + 560)] = kernel[(((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 8) + 70), 3)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 8), 24)*3)) + rx.outer.outer)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ if @tir.likely((threadIdx.x_2 < 96), dtype=bool) {
+ kernel.shared_1[(threadIdx.x_2 + 672)] = kernel[((((((blockIdx.x*147456) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_1) + (floormod(threadIdx.x_2, 24)*3)) + rx.outer.outer) + 129024)]
}
- for (rc.outer.inner: int32, 0, 16) {
- let cse_var_13: int32 = (rc.outer.inner*3)
- let cse_var_12: int32 = (cse_var_13 + 1)
- let cse_var_11: int32 = (cse_var_13 + 144)
- let cse_var_10: int32 = (cse_var_13 + 146)
- let cse_var_9: int32 = (cse_var_13 + 2)
- let cse_var_8: int32 = (cse_var_13 + 48)
- let cse_var_7: int32 = (cse_var_13 + 49)
- let cse_var_6: int32 = (cse_var_13 + 50)
- let cse_var_5: int32 = (cse_var_13 + 96)
- let cse_var_4: int32 = (cse_var_13 + 97)
- let cse_var_3: int32 = (cse_var_13 + 98)
- let cse_var_2: int32 = (cse_var_13 + 145)
- {
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((rc.outer.inner*63) + threadIdx.x)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((rc.outer.inner*63) + threadIdx.x)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 56)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 56)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((rc.outer.inner*63) + threadIdx.x)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[((rc.outer.inner*63) + threadIdx.x)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_10]))
- conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_10]))
- conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_10]))
- conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_10]))
- conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_10]))
- conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_10]))
- conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 56)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 56)]*kernel.shared_1[cse_var_10]))
+ for (rc.outer.inner: int32, 0, 4) {
+ for (ry.outer.inner: int32, 0, 3) {
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7))]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 7)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 7)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 70)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 70)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 14)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 14)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 77)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 77)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 21)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 21)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 84)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 84)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 28)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 28)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 35)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 35)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 98)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 98)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 42)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 42)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 105)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 105)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
}
}
}
}
}
- for (i1.inner: int32, 0, 4) {
+ for (i1.inner: int32, 0, 2) {
for (i2.inner: int32, 0, 7) {
- compute[((((blockIdx.x*196) + (i1.inner*49)) + (i2.inner*7)) + threadIdx.x)] = max((conv2d_nchw_1[((i1.inner*7) + i2.inner)] + bias[((blockIdx.x*4) + i1.inner)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 7)*98)) + (i1.inner*49)) + (i2.inner*7)) + floormod(threadIdx.x, 7))] = max((conv2d_nchw_1[((i1.inner*7) + i2.inner)] + bias[(((blockIdx.x*32) + (floordiv(threadIdx.x, 7)*2)) + i1.inner)]), 0f32)
}
}
}
@@ -764,7 +365,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 0.365 ms
+ Execution time of this operator: 0.413 ms
@@ -809,8 +410,8 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=2)
- conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
- conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=1)
+ conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
+ conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=16)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=7)
@@ -820,18 +421,18 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
- conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=1)
- conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=16)
- conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=3)
- conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
+ conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
+ conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+ conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
+ conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2 [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
- compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=4)
- compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=1)
+ compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
+ compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=16)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=7)
compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
@@ -857,14 +458,14 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=7)
+ kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
- pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=7)
+ pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
- s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 512)
+ s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 64)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
CUDA source code:
@@ -882,10 +483,10 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
#define int64_t long long
#define uint64_t unsigned long long
#endif
- extern "C" __global__ void __launch_bounds__(7) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[28];
- __shared__ float pad_temp_shared[1008];
- __shared__ float kernel_shared[192];
+ extern "C" __global__ void __launch_bounds__(112) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[14];
+ __shared__ float pad_temp_shared[504];
+ __shared__ float kernel_shared[768];
conv2d_nchw[0] = 0.000000e+00f;
conv2d_nchw[7] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
@@ -900,289 +501,63 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
conv2d_nchw[12] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
conv2d_nchw[13] = 0.000000e+00f;
- conv2d_nchw[14] = 0.000000e+00f;
- conv2d_nchw[21] = 0.000000e+00f;
- conv2d_nchw[15] = 0.000000e+00f;
- conv2d_nchw[22] = 0.000000e+00f;
- conv2d_nchw[16] = 0.000000e+00f;
- conv2d_nchw[23] = 0.000000e+00f;
- conv2d_nchw[17] = 0.000000e+00f;
- conv2d_nchw[24] = 0.000000e+00f;
- conv2d_nchw[18] = 0.000000e+00f;
- conv2d_nchw[25] = 0.000000e+00f;
- conv2d_nchw[19] = 0.000000e+00f;
- conv2d_nchw[26] = 0.000000e+00f;
- conv2d_nchw[20] = 0.000000e+00f;
- conv2d_nchw[27] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 32; ++rc_outer_outer) {
+ for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
for (int rx_outer_outer = 0; rx_outer_outer < 3; ++rx_outer_outer) {
__syncthreads();
- pad_temp_shared[((int)threadIdx.x)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 7)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) - 1)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 14)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 6)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 21)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 13)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 28)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 20)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 35)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 27)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 42)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 34)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 49)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 41)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 56)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 63)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 70)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 48)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 77)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 55)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 84)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 62)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 91)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 69)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 98)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 76)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 105)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 83)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 112)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 90)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 119)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 126)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 133)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 97)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 140)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 104)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 147)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 111)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 154)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 118)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 161)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 125)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 168)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 132)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 175)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 139)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 182)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 189)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 196)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 146)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 203)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 153)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 210)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 160)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 217)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 167)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 224)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 174)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 231)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 181)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 238)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 188)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 245)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 252)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 259)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 195)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 266)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 202)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 273)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 209)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 280)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 216)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 287)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 223)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 294)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 230)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 301)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 237)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 308)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 315)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 322)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 244)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 329)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 251)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 336)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 258)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 343)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 265)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 350)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 272)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 357)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 279)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 364)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 286)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 371)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 378)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 385)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 293)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 392)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 300)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 399)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 307)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 406)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 314)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 413)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 321)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 420)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 328)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 427)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 335)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 434)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 441)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 448)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 342)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 455)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 349)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 462)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 356)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 469)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 363)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 476)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 370)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 483)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 377)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 490)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 384)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 497)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 504)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 511)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 391)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 518)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 398)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 525)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 405)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 532)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 412)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 539)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 419)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 546)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 426)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 553)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 433)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 560)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 567)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 574)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 440)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 581)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 447)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 588)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 454)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 595)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 461)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 602)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 468)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 609)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 475)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 616)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 482)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 623)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 630)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 637)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 489)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 644)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 496)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 651)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 503)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 658)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 510)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 665)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 517)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 672)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 524)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 679)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 531)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 686)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 693)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 700)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 538)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 707)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 545)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 714)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 552)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 721)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 559)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 728)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 566)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 735)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 573)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 742)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 580)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 749)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 756)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 763)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 587)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 770)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 594)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 777)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 601)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 784)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 608)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 791)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 615)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 798)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 622)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 805)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 629)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 812)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 819)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 826)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 636)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 833)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 643)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 840)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 650)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 847)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 657)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 854)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 664)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 861)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 671)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 868)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 678)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 875)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 882)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 889)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 685)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 896)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 692)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 903)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 699)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 910)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 706)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 917)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 713)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 924)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 720)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 931)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 727)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 938)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 945)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 952)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 734)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 959)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 741)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 966)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 748)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 973)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 755)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 980)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 762)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 987)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 769)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 994)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 776)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1001)] = 0.000000e+00f;
- kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 18432) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 7)] = kernel[(((((((int)blockIdx.x) * 18432) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 21)];
- kernel_shared[(((int)threadIdx.x) + 14)] = kernel[(((((((int)blockIdx.x) * 18432) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 42)];
- kernel_shared[(((int)threadIdx.x) + 21)] = kernel[(((((((int)blockIdx.x) * 18432) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 63)];
- kernel_shared[(((int)threadIdx.x) + 28)] = kernel[(((((((int)blockIdx.x) * 18432) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 84)];
- kernel_shared[(((int)threadIdx.x) + 35)] = kernel[(((((((int)blockIdx.x) * 18432) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 105)];
- kernel_shared[(((int)threadIdx.x) + 42)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 42) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 42) % 48) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 49)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 49) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 1) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 56)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 56) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 8) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 63)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 63) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 15) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 70)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 70) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 22) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 77)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 77) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 29) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 84)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 84) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 36) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 91)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 91) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 43) % 48) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 98)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 98) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 2) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 105)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 105) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 9) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 112)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 112) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 16) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 119)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 119) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 23) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 126)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 126) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 30) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 133)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 133) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 37) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 140)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 140) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 44) % 48) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 147)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 147) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 3) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 154)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 154) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 10) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 161)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 161) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 17) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 168)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 168) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 24) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 175)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 175) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 31) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 182)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 182) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 38) * 3)) + rx_outer_outer)];
- if (((int)threadIdx.x) < 3) {
- kernel_shared[(((int)threadIdx.x) + 189)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 189) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 45) * 3)) + rx_outer_outer)];
+ pad_temp_shared[((int)threadIdx.x)] = (((((7 <= (((int)threadIdx.x) % 63)) && ((((int)threadIdx.x) % 63) < 56)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 63) * 49)) + rx_outer_outer) + (((int)threadIdx.x) % 63)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 112)] = (((((1 <= (((((int)threadIdx.x) / 7) + 7) % 9)) && ((((((int)threadIdx.x) / 7) + 7) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 112) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 7) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 224) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 336)] = (((((1 <= (((((int)threadIdx.x) / 7) + 3) % 9)) && ((((((int)threadIdx.x) / 7) + 3) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 336) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 3) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 56) {
+ pad_temp_shared[(((int)threadIdx.x) + 448)] = ((((((int)threadIdx.x) < 49) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 448) / 63) * 49)) + (((((int)threadIdx.x) / 7) + 1) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ }
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 112)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 16) % 24) * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 224)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 8) % 24) * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 336)] = kernel[((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3)) + rx_outer_outer) + 64512)];
+ kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 16) % 24) * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 560)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 8) % 24) * 3)) + rx_outer_outer)];
+ if (((int)threadIdx.x) < 96) {
+ kernel_shared[(((int)threadIdx.x) + 672)] = kernel[((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3)) + rx_outer_outer) + 129024)];
}
__syncthreads();
- for (int rc_outer_inner = 0; rc_outer_inner < 16; ++rc_outer_inner) {
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((rc_outer_inner * 63) + ((int)threadIdx.x))] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((rc_outer_inner * 63) + ((int)threadIdx.x))] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 56)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 56)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((rc_outer_inner * 63) + ((int)threadIdx.x))] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((rc_outer_inner * 63) + ((int)threadIdx.x))] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 56)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 56)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
+ for (int rc_outer_inner = 0; rc_outer_inner < 4; ++rc_outer_inner) {
+ for (int ry_outer_inner = 0; ry_outer_inner < 3; ++ry_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 7)] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 7)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 70)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 70)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 14)] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 14)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 77)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 77)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 21)] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 21)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 84)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 84)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 28)] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 28)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 35)] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 35)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 98)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 98)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 42)] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 42)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 105)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 105)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ }
}
}
}
- for (int i1_inner = 0; i1_inner < 4; ++i1_inner) {
+ for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
- compute[((((((int)blockIdx.x) * 196) + (i1_inner * 49)) + (i2_inner * 7)) + ((int)threadIdx.x))] = max((conv2d_nchw[((i1_inner * 7) + i2_inner)] + bias[((((int)blockIdx.x) * 4) + i1_inner)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7))] = max((conv2d_nchw[((i1_inner * 7) + i2_inner)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
}
}
}
@@ -1242,7 +617,7 @@ In the example below we resume the status and do more 5 trials.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 2 minutes 39.840 seconds)
+ **Total running time of the script:** ( 2 minutes 27.183 seconds)
.. _sphx_glr_download_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py:
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
index 62da7f9ee..62a7671a1 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
@@ -614,7 +614,7 @@ so we can read the log file and load the best schedules.
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 9.7388 9.7416 9.7603 9.7146 0.0188
+ 9.9670 9.9606 10.0306 9.9100 0.0494
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
index 5d9408b10..8cf8aa4f4 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
@@ -633,7 +633,7 @@ so we can read the log file and load the best schedules.
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 744.2534 743.2250 747.7378 741.7976 2.5318
+ 755.3439 753.5566 759.6341 752.8411 3.0476
@@ -658,7 +658,7 @@ Other Tips
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 18.913 seconds)
+ **Total running time of the script:** ( 1 minutes 20.913 seconds)
.. _sphx_glr_download_how_to_tune_with_autoscheduler_tune_network_x86.py:
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
index 5037e8077..ebb9d48c4 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
@@ -362,217 +362,28 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
- preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), placeholder_5: placeholder_16: Buffer(placeholder_10, float32, [128, 256], []), placeholder_6: placeholder_17: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_9: placeholder_19: Buffer(placeholder_14, float32, [128, 512], [])} {
- for (i0.outer.i1.outer.fused: int32, 0, 512) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [128]), storage_scope = global {
- for (nb_j.inner: int32, 0, 2) {
- let cse_var_2: int32 = (nb_j.inner*16)
- let cse_var_1: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
- {
- compute_5: Buffer(compute_4, float32, [128], [])[cse_var_2] = 0f32
- compute_5[(cse_var_2 + 1)] = 0f32
- compute_5[(cse_var_2 + 2)] = 0f32
- compute_5[(cse_var_2 + 3)] = 0f32
- compute_5[(cse_var_2 + 4)] = 0f32
- compute_5[(cse_var_2 + 5)] = 0f32
- compute_5[(cse_var_2 + 6)] = 0f32
- compute_5[(cse_var_2 + 7)] = 0f32
- compute_5[(cse_var_2 + 8)] = 0f32
- compute_5[(cse_var_2 + 9)] = 0f32
- compute_5[(cse_var_2 + 10)] = 0f32
- compute_5[(cse_var_2 + 11)] = 0f32
- compute_5[(cse_var_2 + 12)] = 0f32
- compute_5[(cse_var_2 + 13)] = 0f32
- compute_5[(cse_var_2 + 14)] = 0f32
- compute_5[(cse_var_2 + 15)] = 0f32
- compute_5[(cse_var_2 + 32)] = 0f32
- compute_5[(cse_var_2 + 33)] = 0f32
- compute_5[(cse_var_2 + 34)] = 0f32
- compute_5[(cse_var_2 + 35)] = 0f32
- compute_5[(cse_var_2 + 36)] = 0f32
- compute_5[(cse_var_2 + 37)] = 0f32
- compute_5[(cse_var_2 + 38)] = 0f32
- compute_5[(cse_var_2 + 39)] = 0f32
- compute_5[(cse_var_2 + 40)] = 0f32
- compute_5[(cse_var_2 + 41)] = 0f32
- compute_5[(cse_var_2 + 42)] = 0f32
- compute_5[(cse_var_2 + 43)] = 0f32
- compute_5[(cse_var_2 + 44)] = 0f32
- compute_5[(cse_var_2 + 45)] = 0f32
- compute_5[(cse_var_2 + 46)] = 0f32
- compute_5[(cse_var_2 + 47)] = 0f32
- compute_5[(cse_var_2 + 64)] = 0f32
- compute_5[(cse_var_2 + 65)] = 0f32
- compute_5[(cse_var_2 + 66)] = 0f32
- compute_5[(cse_var_2 + 67)] = 0f32
- compute_5[(cse_var_2 + 68)] = 0f32
- compute_5[(cse_var_2 + 69)] = 0f32
- compute_5[(cse_var_2 + 70)] = 0f32
- compute_5[(cse_var_2 + 71)] = 0f32
- compute_5[(cse_var_2 + 72)] = 0f32
- compute_5[(cse_var_2 + 73)] = 0f32
- compute_5[(cse_var_2 + 74)] = 0f32
- compute_5[(cse_var_2 + 75)] = 0f32
- compute_5[(cse_var_2 + 76)] = 0f32
- compute_5[(cse_var_2 + 77)] = 0f32
- compute_5[(cse_var_2 + 78)] = 0f32
- compute_5[(cse_var_2 + 79)] = 0f32
- compute_5[(cse_var_2 + 96)] = 0f32
- compute_5[(cse_var_2 + 97)] = 0f32
- compute_5[(cse_var_2 + 98)] = 0f32
- compute_5[(cse_var_2 + 99)] = 0f32
- compute_5[(cse_var_2 + 100)] = 0f32
- compute_5[(cse_var_2 + 101)] = 0f32
- compute_5[(cse_var_2 + 102)] = 0f32
- compute_5[(cse_var_2 + 103)] = 0f32
- compute_5[(cse_var_2 + 104)] = 0f32
- compute_5[(cse_var_2 + 105)] = 0f32
- compute_5[(cse_var_2 + 106)] = 0f32
- compute_5[(cse_var_2 + 107)] = 0f32
- compute_5[(cse_var_2 + 108)] = 0f32
- compute_5[(cse_var_2 + 109)] = 0f32
- compute_5[(cse_var_2 + 110)] = 0f32
- compute_5[(cse_var_2 + 111)] = 0f32
- for (elem_idx: int32, 0, (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
- let cse_var_67: int32 = (cse_var_2 + 41)
- let cse_var_66: int32 = (cse_var_2 + 40)
- let cse_var_65: int32 = (cse_var_2 + 4)
- let cse_var_64: int32 = (cse_var_2 + 39)
- let cse_var_63: int32 = (cse_var_2 + 38)
- let cse_var_62: int32 = (cse_var_2 + 37)
- let cse_var_61: int32 = (cse_var_2 + 36)
- let cse_var_60: int32 = (cse_var_2 + 35)
- let cse_var_59: int32 = (cse_var_2 + 34)
- let cse_var_58: int32 = (cse_var_2 + 33)
- let cse_var_57: int32 = (cse_var_2 + 32)
- let cse_var_56: int32 = (cse_var_2 + 3)
- let cse_var_55: int32 = (cse_var_2 + 2)
- let cse_var_54: int32 = (cse_var_2 + 15)
- let cse_var_53: int32 = (cse_var_2 + 14)
- let cse_var_52: int32 = (cse_var_2 + 1)
- let cse_var_51: int32 = (cse_var_2 + 12)
- let cse_var_50: int32 = (cse_var_2 + 111)
- let cse_var_49: int32 = (cse_var_2 + 110)
- let cse_var_48: int32 = (cse_var_2 + 11)
- let cse_var_47: int32 = (cse_var_2 + 109)
- let cse_var_46: int32 = (cse_var_2 + 108)
- let cse_var_45: int32 = (cse_var_2 + 107)
- let cse_var_44: int32 = (cse_var_2 + 106)
- let cse_var_43: int32 = (cse_var_2 + 105)
- let cse_var_42: int32 = (cse_var_2 + 104)
- let cse_var_41: int32 = (cse_var_2 + 103)
- let cse_var_40: int32 = (cse_var_2 + 102)
- let cse_var_39: int32 = (cse_var_2 + 101)
- let cse_var_38: int32 = (cse_var_2 + 100)
- let cse_var_37: int32 = (cse_var_2 + 10)
- let cse_var_36: int32 = (cse_var_2 + 13)
- let cse_var_35: int32 = (elem_idx*16)
- let cse_var_34: int32 = (cse_var_2 + 99)
- let cse_var_33: int32 = (cse_var_2 + 98)
- let cse_var_32: int32 = (cse_var_2 + 97)
- let cse_var_31: int32 = (cse_var_2 + 96)
- let cse_var_30: int32 = (cse_var_2 + 9)
- let cse_var_29: int32 = (cse_var_2 + 8)
- let cse_var_28: int32 = (cse_var_2 + 79)
- let cse_var_27: int32 = (cse_var_2 + 78)
- let cse_var_26: int32 = (cse_var_2 + 77)
- let cse_var_25: int32 = (cse_var_2 + 76)
- let cse_var_24: int32 = (cse_var_2 + 75)
- let cse_var_23: int32 = (cse_var_2 + 74)
- let cse_var_22: int32 = (cse_var_2 + 73)
- let cse_var_21: int32 = (cse_var_2 + 72)
- let cse_var_20: int32 = (cse_var_2 + 71)
- let cse_var_19: int32 = (cse_var_2 + 42)
- let cse_var_18: int32 = (cse_var_2 + 44)
- let cse_var_17: int32 = (cse_var_2 + 45)
- let cse_var_16: int32 = (cse_var_2 + 46)
- let cse_var_15: int32 = (cse_var_2 + 47)
- let cse_var_14: int32 = (cse_var_2 + 5)
- let cse_var_13: int32 = (cse_var_2 + 6)
- let cse_var_12: int32 = (cse_var_2 + 64)
- let cse_var_11: int32 = (cse_var_2 + 65)
- let cse_var_10: int32 = (cse_var_2 + 66)
- let cse_var_9: int32 = (cse_var_2 + 67)
- let cse_var_8: int32 = (cse_var_2 + 68)
- let cse_var_7: int32 = (cse_var_2 + 69)
- let cse_var_6: int32 = (cse_var_2 + 7)
- let cse_var_5: int32 = (cse_var_2 + 43)
- let cse_var_4: int32 = (cse_var_2 + 70)
- let cse_var_3: int32 = (floordiv(i0.outer.i1.outer.fused, 16)*1024)
- {
- compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_35)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_52] = (compute_5[cse_var_52] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 1)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_55] = (compute_5[cse_var_55] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 2)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_56] = (compute_5[cse_var_56] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 3)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_65] = (compute_5[cse_var_65] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 4)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 5)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 6)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 7)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_29] = (compute_5[cse_var_29] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 8)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_30] = (compute_5[cse_var_30] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 9)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_37] = (compute_5[cse_var_37] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 10)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_48] = (compute_5[cse_var_48] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 11)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_51] = (compute_5[cse_var_51] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 12)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_36] = (compute_5[cse_var_36] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 13)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_53] = (compute_5[cse_var_53] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 14)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_54] = (compute_5[cse_var_54] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 15)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_57] = (compute_5[cse_var_57] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_35)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_58] = (compute_5[cse_var_58] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_59] = (compute_5[cse_var_59] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_60] = (compute_5[cse_var_60] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_61] = (compute_5[cse_var_61] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_62] = (compute_5[cse_var_62] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_63] = (compute_5[cse_var_63] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_64] = (compute_5[cse_var_64] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_66] = (compute_5[cse_var_66] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_67] = (compute_5[cse_var_67] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_19] = (compute_5[cse_var_19] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_18] = (compute_5[cse_var_18] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_35)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_20] = (compute_5[cse_var_20] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_21] = (compute_5[cse_var_21] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_22] = (compute_5[cse_var_22] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_23] = (compute_5[cse_var_23] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_24] = (compute_5[cse_var_24] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_25] = (compute_5[cse_var_25] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_26] = (compute_5[cse_var_26] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_27] = (compute_5[cse_var_27] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_28] = (compute_5[cse_var_28] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_31] = (compute_5[cse_var_31] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_35)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_32] = (compute_5[cse_var_32] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_33] = (compute_5[cse_var_33] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_34] = (compute_5[cse_var_34] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_38] = (compute_5[cse_var_38] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_39] = (compute_5[cse_var_39] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_40] = (compute_5[cse_var_40] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_41] = (compute_5[cse_var_41] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_42] = (compute_5[cse_var_42] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_43] = (compute_5[cse_var_43] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_44] = (compute_5[cse_var_44] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_45] = (compute_5[cse_var_45] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_46] = (compute_5[cse_var_46] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_47] = (compute_5[cse_var_47] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_49] = (compute_5[cse_var_49] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_50] = (compute_5[cse_var_50] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ preflattened_buffer_map = {placeholder_8: placeholder_15: Buffer(placeholder_13, int32, [33], []), placeholder_5: placeholder_16: Buffer(placeholder_10, float32, [128, 256], []), placeholder_9: placeholder_17: Buffer(placeholder_14, float32, [128, 512], []), placeholder_7: placeholder_18: Buffer(placeholder_12, int32, [4916], []), placeholder_6: placeholder_19: Buffer(placeholder_11, float32, [4916, 16, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], [])} {
+ for (i0.outer.i1.outer.fused: int32, 0, 128) "parallel" {
+ allocate(compute_4: Pointer(global float32), float32, [512]), storage_scope = global {
+ for (i.outer.inner: int32, 0, 8) {
+ for (i.inner.init: int32, 0, 4) {
+ for (j.init: int32, 0, 16) {
+ compute_5: Buffer(compute_4, float32, [512], [])[(((i.outer.inner*64) + (i.inner.init*16)) + j.init)] = 0f32
+ }
+ }
+ for (elem_idx: int32, 0, let cse_var_1: int32 = floormod(i0.outer.i1.outer.fused, 32) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
+ for (i.inner: int32, 0, 4) {
+ for (j: int32, 0, 16) {
+ let cse_var_3: int32 = floormod(i0.outer.i1.outer.fused, 32)
+ let cse_var_2: int32 = (((i.outer.inner*64) + (i.inner*16)) + j)
+ compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[((((floordiv(i0.outer.i1.outer.fused, 32)*8192) + (i.outer.inner*1024)) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
}
}
}
}
- for (i0.inner: int32, 0, 4) {
- for (i1.inner: int32, 0, 32) {
- let cse_var_68: int32 = ((((floordiv(i0.outer.i1.outer.fused, 16)*2048) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32)) + i1.inner)
- compute[cse_var_68] = max((compute_5[((i0.inner*32) + i1.inner)] + placeholder_4[cse_var_68]), 0f32)
- }
+ for (i0.inner: int32, 0, 32) {
+ let cse_var_4: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*16384) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 32)*16))
+ compute[ramp(cse_var_4, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_4, 1, 16)]), broadcast(0f32, 16))
}
}
}
@@ -626,7 +437,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 2.587 ms
+ Execution time of this operator: 1.474 ms
diff --git a/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt b/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
index c201ceea1..9565b9f52 100644
--- a/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
Computation times
=================
-**00:44.049** total execution time for **how_to_tune_with_autotvm** files:
+**00:44.195** total execution time for **how_to_tune_with_autotvm** files:
-- **00:43.190**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``)
-- **00:00.224**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``)
-- **00:00.213**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``)
-- **00:00.211**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``)
-- **00:00.211**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``)
+- **00:43.373**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``)
+- **00:00.217**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``)
+- **00:00.205**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``)
+- **00:00.200**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``)
+- **00:00.200**: :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``)
diff --git a/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt b/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
index d17af97c4..dcaee000d 100644
--- a/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
@@ -859,8 +859,8 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2885496
- No: 6 GFLOPS: 42.26/42.26 result: MeasureResult(costs=(0.0054775518947368425,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5810227394104004, timestamp=1650670409.569581) [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
- No: 7 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+ No: 6 GFLOPS: 102.59/102.59 result: MeasureResult(costs=(0.0022566282291666667,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5949983596801758, timestamp=1650690027.666691) [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
+ No: 7 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -983,7 +983,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 16, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6225319
- No: 8 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+ No: 8 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1106,7 +1106,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,943546
- No: 9 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+ No: 9 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1229,7 +1229,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 16, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2868708
- No: 10 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+ No: 10 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 142, in build
res = future.result()
File "/usr/lib/python3.7/concurrent/futures/_base.py", line 435, in result
@@ -1247,7 +1247,7 @@ for this template
TimeoutError
[('tile_f', [-1, 32, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4691833
- No: 11 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+ No: 11 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1370,7 +1370,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 64]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1042124
- No: 12 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+ No: 12 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1493,7 +1493,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10013405
- No: 13 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+ No: 13 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1616,7 +1616,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6732082
- No: 14 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+ No: 14 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1739,7 +1739,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 4, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7536735
- No: 15 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+ No: 15 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1862,7 +1862,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,482121
- No: 16 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+ No: 16 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1985,7 +1985,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2824525
- No: 17 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+ No: 17 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2108,7 +2108,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4559286
- No: 18 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+ No: 18 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2231,7 +2231,7 @@ for this template
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9677544
- No: 19 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+ No: 19 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 721, in __call__
yield remote, remote.load_module(os.path.split(build_result.filename)[1])
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 685, in run_through_rpc
@@ -2319,7 +2319,7 @@ for this template
15: _PyEval_EvalFrameDefault
14: 0x0000000000537c30
13: _PyObject_FastCallKeywords
- 12: 0x00007fce6fd74fa2
+ 12: 0x00007f48339cefa2
11: _ctypes_callproc
10: ffi_call
9: ffi_call_unix64
@@ -2384,7 +2384,7 @@ for this template
21: _PyFunction_FastCallKeywords
20: _PyEval_EvalFrameDefault
19: _PyFunction_FastCall [('tile_f', [-1, 8, 2, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6390073
- No: 20 GFLOPS: 144.55/144.55 result: MeasureResult(costs=(0.0016015395100000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3942515850067139, timestamp=1650670435.7232537) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
+ No: 20 GFLOPS: 142.38/142.38 result: MeasureResult(costs=(0.0016259662,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.420372486114502, timestamp=1650690054.0953064) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
@@ -2437,7 +2437,7 @@ and measure running time.
Best config:
[('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
- Time cost of this operator: 0.001962
+ Time cost of this operator: 0.001982
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
index d1244112a..34f54accf 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
@@ -292,10 +292,10 @@ Timing the untuned program
########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs
--------- --- -------- ------- ----- ------ -------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 305.2 98.718 (1, 2, 10, 10, 3) 2 1
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.018 0.976 (1, 6, 10, 10) 1 1
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.945 0.306 (1, 1, 10, 10, 3) 1 1
- Total_time - 309.163 - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 312.8 98.738 (1, 2, 10, 10, 3) 2 1
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.073 0.97 (1, 6, 10, 10) 1 1
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.925 0.292 (1, 1, 10, 10, 3) 1 1
+ Total_time - 316.798 - - - -
@@ -357,10 +357,10 @@ Timing the tuned program
########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs
--------- --- -------- ------- ----- ------ -------
- tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 150.5 98.203 (1, 6, 10, 10, 1) 2 1
- tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.805 1.178 (1, 6, 10, 10) 1 1
- tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.948 0.619 (1, 1, 10, 10, 3) 1 1
- Total_time - 153.253 - - - -
+ tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 81.0 96.817 (1, 6, 10, 10, 1) 2 1
+ tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.728 2.065 (1, 6, 10, 10) 1 1
+ tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.936 1.118 (1, 1, 10, 10, 3) 1 1
+ Total_time - 83.663 - - - -
diff --git a/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
index 6b203db54..c12a6f7e5 100644
--- a/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
Computation times
=================
-**00:43.963** total execution time for **how_to_work_with_microtvm** files:
+**00:44.000** total execution time for **how_to_work_with_microtvm** files:
-- **00:39.912**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)
-- **00:03.462**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)
-- **00:00.199**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tvmc.py` (``micro_tvmc.py``)
-- **00:00.197**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_reference_vm.py` (``micro_reference_vm.py``)
-- **00:00.193**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``)
+- **00:39.918**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)
+- **00:03.494**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)
+- **00:00.208**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_reference_vm.py` (``micro_reference_vm.py``)
+- **00:00.191**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``)
+- **00:00.190**: :ref:`sphx_glr_how_to_work_with_microtvm_micro_tvmc.py` (``micro_tvmc.py``)
diff --git a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
index 7fc41e479..ebc8fb17d 100644
--- a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
@@ -5,8 +5,8 @@
Computation times
=================
-**00:09.266** total execution time for **how_to_work_with_relay** files:
+**00:09.154** total execution time for **how_to_work_with_relay** files:
-- **00:07.223**: :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)
-- **00:01.830**: :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)
+- **00:07.437**: :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)
+- **00:01.503**: :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)
- **00:00.214**: :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``)
diff --git a/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
index 79bffc008..ad328af6b 100644
--- a/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
@@ -5,13 +5,13 @@
Computation times
=================
-**00:05.515** total execution time for **how_to_work_with_schedules** files:
+**00:05.549** total execution time for **how_to_work_with_schedules** files:
-- **00:02.014**: :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)
-- **00:01.101**: :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)
-- **00:00.712**: :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)
-- **00:00.692**: :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)
-- **00:00.304**: :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)
-- **00:00.237**: :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``)
-- **00:00.236**: :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``)
-- **00:00.218**: :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``)
+- **00:02.037**: :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)
+- **00:01.145**: :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)
+- **00:00.713**: :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)
+- **00:00.690**: :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)
+- **00:00.298**: :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)
+- **00:00.227**: :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``)
+- **00:00.224**: :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``)
+- **00:00.217**: :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``)
diff --git a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
index 9fb2359e7..3411e7fa0 100644
--- a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
@@ -318,7 +318,7 @@ The importing needs to happen before the tensorized GEMV being executed.
C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
buffer_map = {A_1: A, B_1: B, C_1: C}
preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpqgpys2ao/input0.cc'\nsource_filename = \"/tmp/tmpqgpys2ao/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = alloca float*, align 8\n %8 = alloca float*, align 8\n %9 = alloca floa [...]
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmp6yusbrem/input0.cc'\nsource_filename = \"/tmp/tmp6yusbrem/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = alloca float*, align 8\n %8 = alloca float*, align 8\n %9 = alloca floa [...]
for (i, 0, 1024) {
for (j.outer: int32, 0, 32) {
@tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
diff --git a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
index 1c266fa69..44e73c730 100644
--- a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
@@ -5,7 +5,7 @@
Computation times
=================
-**00:20.608** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:20.754** total execution time for **topic_vta_tutorials_autotvm** files:
-- **00:20.405**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``)
-- **00:00.203**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``)
+- **00:20.557**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``)
+- **00:00.197**: :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``)
diff --git a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
index 8696660ba..1e324867f 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -265,7 +265,7 @@ The compilation steps are:
DeprecationWarning,
/workspace/vta/tutorials/frontend/deploy_classification.py:213: DeprecationWarning: legacy graph executor behavior of producing json / lib / params will be removed in the next release. Please see documents of tvm.contrib.graph_executor.GraphModule for the new recommended usage.
relay_prog, target=tvm.target.Target(target, host=env.target_host), params=params
- resnet18_v1 inference graph built in 21.57s!
+ resnet18_v1 inference graph built in 20.94s!
diff --git a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
index 60037fcc5..132beddbf 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -301,7 +301,7 @@ The compilation steps are:
/workspace/python/tvm/relay/build_module.py:439: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
- yolov3-tiny inference graph built in 14.97s!
+ yolov3-tiny inference graph built in 14.56s!
diff --git a/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt
index 9d6699734..529dd8262 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt
@@ -5,7 +5,7 @@
Computation times
=================
-**01:27.486** total execution time for **topic_vta_tutorials_frontend** files:
+**01:27.989** total execution time for **topic_vta_tutorials_frontend** files:
-- **00:46.206**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)
-- **00:41.280**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``)
+- **00:46.818**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)
+- **00:41.170**: :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``)
diff --git a/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
index 0e675dddd..b0fbf5c43 100644
--- a/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
@@ -5,7 +5,7 @@
Computation times
=================
-**00:03.472** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.767** total execution time for **topic_vta_tutorials_optimize** files:
-- **00:02.936**: :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)
-- **00:00.537**: :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``)
+- **00:03.229**: :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)
+- **00:00.538**: :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``)
diff --git a/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
index d1651369a..815b9cd7a 100644
--- a/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
@@ -5,7 +5,7 @@
Computation times
=================
-**00:00.971** total execution time for **topic_vta_tutorials** files:
+**00:00.976** total execution time for **topic_vta_tutorials** files:
-- **00:00.494**: :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``)
-- **00:00.476**: :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``)
+- **00:00.499**: :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``)
+- **00:00.477**: :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``)
diff --git a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
index 156b7a7ed..6c79e7090 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -306,7 +306,7 @@ We build the binary and check its correctness and performance.
.. code-block:: none
- Execution time of this operator: 93.500 ms
+ Execution time of this operator: 93.729 ms
@@ -402,7 +402,7 @@ resume the status and do more 5 trials.
Resume search:
/usr/local/lib/python3.7/dist-packages/xgboost/training.py:17: UserWarning: Old style callback is deprecated. See: https://xgboost.readthedocs.io/en/latest/python/callbacks.html
warnings.warn(f'Old style callback is deprecated. See: {link}', UserWarning)
- *E
+
@@ -417,7 +417,7 @@ operations.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 1 minutes 7.232 seconds)
+ **Total running time of the script:** ( 1 minutes 3.810 seconds)
.. _sphx_glr_download_tutorial_auto_scheduler_matmul_x86.py:
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index 758a57c97..202fbd570 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -268,7 +268,7 @@ standard deviation.
.. code-block:: none
- {'mean': 496.60638313000044, 'median': 496.4213516499967, 'std': 0.8980274498640965}
+ {'mean': 492.6316053099993, 'median': 492.4662148999971, 'std': 0.45000288160307683}
@@ -482,31 +482,32 @@ the tuning data to.
.. code-block:: none
-
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 1/25] Current/Best: 9.89/ 12.02 GFLOPS | Progress: (4/10) | 6.34 s
[Task 1/25] Current/Best: 8.53/ 23.62 GFLOPS | Progress: (8/10) | 8.48 s
[Task 1/25] Current/Best: 10.63/ 23.62 GFLOPS | Progress: (10/10) | 9.56 s Done.
-
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 2/25] Current/Best: 7.44/ 20.07 GFLOPS | Progress: (4/10) | 2.11 s
[Task 2/25] Current/Best: 15.55/ 20.07 GFLOPS | Progress: (8/10) | 3.19 s
[Task 2/25] Current/Best: 22.68/ 22.68 GFLOPS | Progress: (10/10) | 3.62 s Done.
-
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 3/25] Current/Best: 24.13/ 24.13 GFLOPS | Progress: (4/10) | 2.99 s
[Task 3/25] Current/Best: 12.03/ 24.13 GFLOPS | Progress: (8/10) | 7.25 s
[Task 3/25] Current/Best: 16.98/ 24.13 GFLOPS | Progress: (10/10) | 8.16 s Done.
-
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 4/25] Current/Best: 14.55/ 17.66 GFLOPS | Progress: (4/10) | 3.01 s
[Task 4/25] Current/Best: 10.05/ 17.66 GFLOPS | Progress: (8/10) | 4.51 s
[Task 4/25] Current/Best: 16.66/ 17.66 GFLOPS | Progress: (10/10) | 5.20 s Done.
-
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 5/25] Current/Best: 1.71/ 18.05 GFLOPS | Progress: (4/10) | 4.45 s
[Task 5/25] Current/Best: 9.74/ 18.05 GFLOPS | Progress: (8/10) | 6.92 s
[Task 5/25] Current/Best: 18.35/ 18.35 GFLOPS | Progress: (10/10) | 7.64 s Done.
-
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 6/25] Current/Best: 9.62/ 21.76 GFLOPS | Progress: (4/10) | 3.03 s
[Task 6/25] Current/Best: 16.67/ 21.76 GFLOPS | Progress: (8/10) | 4.75 s
[Task 6/25] Current/Best: 13.17/ 21.76 GFLOPS | Progress: (10/10) | 6.49 s Done.
-
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 7/25] Current/Best: 12.34/ 15.54 GFLOPS | Progress: (4/10) | 3.80 s
[Task 7/25] Current/Best: 9.77/ 17.94 GFLOPS | Progress: (8/10) | 6.40 s
[Task 7/25] Current/Best: 1.58/ 17.94 GFLOPS | Progress: (10/10) | 8.66 s Done.
-
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 8/25] Current/Best: 19.10/ 19.10 GFLOPS | Progress: (4/10) | 3.08 s
[Task 8/25] Current/Best: 10.45/ 19.10 GFLOPS | Progress: (8/10) | 5.11 s
[Task 8/25] Current/Best: 10.46/ 19.10 GFLOPS | Progress: (10/10) | 6.74 s Done.
-
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 9/25] Current/Best: 17.12/ 22.42 GFLOPS | Progress: (4/10) | 2.43 s
[Task 9/25] Current/Best: 15.09/ 22.42 GFLOPS | Progress: (8/10) | 3.85 s
[Task 9/25] Current/Best: 20.70/ 22.42 GFLOPS | Progress: (10/10) | 6.31 s Done.
-
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 10/25] Current/Best: 13.05/ 13.05 GFLOPS | Progress: (4/10) | 2.34 s
[Task 10/25] Current/Best: 9.54/ 20.94 GFLOPS | Progress: (8/10) | 3.91 s
[Task 10/25] Current/Best: 16.56/ 20.94 GFLOPS | Progress: (10/10) | 4.81 s Done.
-
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 11/25] Current/Best: 14.02/ 21.31 GFLOPS | Progress: (4/10) | 3.29 s
[Task 11/25] Current/Best: 18.81/ 23.74 GFLOPS | Progress: (8/10) | 5.37 s
[Task 11/25] Current/Best: 19.17/ 23.74 GFLOPS | Progress: (10/10) | 6.23 s Done.
-
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 12/25] Current/Best: 16.29/ 16.29 GFLOPS | Progress: (4/10) | 3.15 s
[Task 12/25] Current/Best: 14.14/ 22.82 GFLOPS | Progress: (8/10) | 4.75 s
[Task 12/25] Current/Best: 8.12/ 22.82 GFLOPS | Progress: (10/10) | 6.28 s Done.
-
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 13/25] Current/Best: 7.80/ 13.32 GFLOPS | Progress: (4/10) | 4.27 s
[Task 13/25] Current/Best: 23.80/ 23.89 GFLOPS | Progress: (8/10) | 6.90 s
[Task 13/25] Current/Best: 16.79/ 23.89 GFLOPS | Progress: (10/10) | 7.80 s Done.
-
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 14/25] Current/Best: 13.94/ 15.06 GFLOPS | Progress: (4/10) | 3.43 s
[Task 14/25] Current/Best: 16.27/ 16.27 GFLOPS | Progress: (8/10) | 4.95 s
[Task 14/25] Current/Best: 10.95/ 16.27 GFLOPS | Progress: (10/10) | 6.05 s
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 15/25] Current/Best: 13.55/ 13.55 GFLOPS | Progress: (4/10) | 3.07 s
[Task 15/25] Current/Best: 3.17/ 13.55 GFLOPS | Progress: (8/10) | 6.84 s
[Task 15/25] Current/Best: 20.67/ 20.67 GFLOPS | Progress: (10/10) | 9.12 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s Done.
+
[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 1/25] Current/Best: 17.15/ 23.91 GFLOPS | Progress: (4/10) | 4.81 s
[Task 1/25] Current/Best: 5.81/ 23.91 GFLOPS | Progress: (8/10) | 8.61 s
[Task 1/25] Current/Best: 16.74/ 23.91 GFLOPS | Progress: (10/10) | 9.73 s Done.
+
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 2/25] Current/Best: 17.55/ 17.55 GFLOPS | Progress: (4/10) | 2.70 s
[Task 2/25] Current/Best: 14.91/ 20.17 GFLOPS | Progress: (8/10) | 3.87 s
[Task 2/25] Current/Best: 20.68/ 20.68 GFLOPS | Progress: (10/10) | 4.35 s Done.
+
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 3/25] Current/Best: 17.86/ 18.99 GFLOPS | Progress: (4/10) | 2.68 s
[Task 3/25] Current/Best: 7.63/ 22.54 GFLOPS | Progress: (8/10) | 4.62 s
[Task 3/25] Current/Best: 14.78/ 22.54 GFLOPS | Progress: (10/10) | 5.56 s Done.
+
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 4/25] Current/Best: 11.82/ 14.62 GFLOPS | Progress: (4/10) | 5.69 s
[Task 4/25] Current/Best: 13.01/ 14.71 GFLOPS | Progress: (8/10) | 7.52 s
[Task 4/25] Current/Best: 14.33/ 16.94 GFLOPS | Progress: (10/10) | 8.30 s Done.
+
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 5/25] Current/Best: 11.36/ 16.91 GFLOPS | Progress: (4/10) | 2.63 s
[Task 5/25] Current/Best: 5.16/ 19.29 GFLOPS | Progress: (8/10) | 4.47 s
[Task 5/25] Current/Best: 9.76/ 19.29 GFLOPS | Progress: (10/10) | 5.14 s Done.
+
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 6/25] Current/Best: 16.33/ 16.33 GFLOPS | Progress: (4/10) | 3.11 s
[Task 6/25] Current/Best: 10.78/ 16.33 GFLOPS | Progress: (8/10) | 6.33 s
[Task 6/25] Current/Best: 13.49/ 18.21 GFLOPS | Progress: (10/10) | 7.05 s Done.
+
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 7/25] Current/Best: 10.59/ 21.24 GFLOPS | Progress: (4/10) | 2.67 s
[Task 7/25] Current/Best: 6.48/ 21.24 GFLOPS | Progress: (8/10) | 4.80 s
[Task 7/25] Current/Best: 6.72/ 21.24 GFLOPS | Progress: (10/10) | 5.96 s Done.
+
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 8/25] Current/Best: 12.71/ 22.83 GFLOPS | Progress: (4/10) | 2.95 s
[Task 8/25] Current/Best: 8.97/ 22.83 GFLOPS | Progress: (8/10) | 5.86 s
[Task 8/25] Current/Best: 5.73/ 22.83 GFLOPS | Progress: (10/10) | 6.93 s Done.
+
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 9/25] Current/Best: 9.99/ 20.35 GFLOPS | Progress: (4/10) | 4.60 s
[Task 9/25] Current/Best: 11.69/ 20.35 GFLOPS | Progress: (8/10) | 6.14 s
[Task 9/25] Current/Best: 12.63/ 20.35 GFLOPS | Progress: (10/10) | 15.62 s Done.
+
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 10/25] Current/Best: 14.13/ 20.33 GFLOPS | Progress: (4/10) | 2.35 s
[Task 10/25] Current/Best: 19.04/ 20.33 GFLOPS | Progress: (8/10) | 4.01 s
[Task 10/25] Current/Best: 19.21/ 20.33 GFLOPS | Progress: (10/10) | 5.13 s Done.
+
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 11/25] Current/Best: 18.01/ 18.01 GFLOPS | Progress: (4/10) | 3.35 s
[Task 11/25] Current/Best: 12.37/ 20.88 GFLOPS | Progress: (8/10) | 5.39 s
[Task 11/25] Current/Best: 1.59/ 20.88 GFLOPS | Progress: (10/10) | 7.66 s Done.
+
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 12/25] Current/Best: 10.73/ 20.18 GFLOPS | Progress: (4/10) | 2.98 s
[Task 12/25] Current/Best: 13.21/ 20.18 GFLOPS | Progress: (8/10) | 6.33 s
[Task 12/25] Current/Best: 16.36/ 20.18 GFLOPS | Progress: (10/10) | 7.35 s Done.
+
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 13/25] Current/Best: 11.47/ 11.47 GFLOPS | Progress: (4/10) | 5.08 s
[Task 13/25] Current/Best: 4.74/ 15.17 GFLOPS | Progress: (8/10) | 8.45 s
[Task 13/25] Current/Best: 18.73/ 18.73 GFLOPS | Progress: (10/10) | 10.62 s Done.
+
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 14/25] Current/Best: 14.51/ 16.34 GFLOPS | Progress: (4/10) | 3.94 s
[Task 14/25] Current/Best: 7.15/ 16.34 GFLOPS | Progress: (8/10) | 7.10 s
[Task 14/25] Current/Best: 13.16/ 19.19 GFLOPS | Progress: (10/10) | 7.86 s Done.
+
[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 15/25] Current/Best: 10.18/ 23.21 GFLOPS | Progress: (4/10) | 2.91 s
[Task 15/25] Current/Best: 17.88/ 23.21 GFLOPS | Progress: (8/10) | 4.53 s
[Task 15/25] Current/Best: 18.97/ 23.21 GFLOPS | Progress: (10/10) | 5.31 s
[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 16/25] Current/Best: 15.88/ 15.88 GFLOPS | Progress: (4/10) | 2.91 s
[Task 16/25] Current/Best: 6.17/ 15.88 GFLOPS | Progress: (8/10) | 4.91 s
[Task 16/25] Current/Best: 10.18/ 15.88 GFLOPS | Progress: (10/10) | 7.78 s Done.
+
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 17/25] Current/Best: 20.55/ 20.55 GFLOPS | Progress: (4/10) | 3.11 s
[Task 17/25] Current/Best: 10.25/ 20.55 GFLOPS | Progress: (8/10) | 6.04 s
[Task 17/25] Current/Best: 7.50/ 20.55 GFLOPS | Progress: (10/10) | 8.26 s Done.
+
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 18/25] Current/Best: 9.55/ 19.41 GFLOPS | Progress: (4/10) | 3.16 s
[Task 18/25] Current/Best: 17.21/ 22.12 GFLOPS | Progress: (8/10) | 4.74 s
[Task 18/25] Current/Best: 16.47/ 22.12 GFLOPS | Progress: (10/10) | 5.95 s Done.
+
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 19/25] Current/Best: 10.75/ 21.84 GFLOPS | Progress: (4/10) | 3.63 s
[Task 19/25] Current/Best: 5.30/ 21.84 GFLOPS | Progress: (8/10) | 7.32 s
[Task 19/25] Current/Best: 22.77/ 22.77 GFLOPS | Progress: (10/10) | 8.56 s Done.
+
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 20/25] Current/Best: 5.34/ 16.10 GFLOPS | Progress: (4/10) | 3.22 s
[Task 20/25] Current/Best: 8.96/ 16.10 GFLOPS | Progress: (8/10) | 4.64 s
[Task 20/25] Current/Best: 5.19/ 16.10 GFLOPS | Progress: (10/10) | 9.68 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s Done.
Done.
-
[Task 16/25] Current/Best: 18.23/ 18.23 GFLOPS | Progress: (4/10) | 2.45 s
[Task 16/25] Current/Best: 10.45/ 18.23 GFLOPS | Progress: (8/10) | 4.22 s
[Task 16/25] Current/Best: 18.05/ 18.23 GFLOPS | Progress: (10/10) | 4.96 s Done.
-
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 17/25] Current/Best: 6.22/ 22.92 GFLOPS | Progress: (4/10) | 2.75 s
[Task 17/25] Current/Best: 6.15/ 22.92 GFLOPS | Progress: (8/10) | 5.02 s
[Task 17/25] Current/Best: 10.99/ 22.92 GFLOPS | Progress: (10/10) | 6.45 s Done.
-
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 18/25] Current/Best: 7.75/ 20.47 GFLOPS | Progress: (4/10) | 3.10 s
[Task 18/25] Current/Best: 19.60/ 20.47 GFLOPS | Progress: (8/10) | 4.72 s
[Task 18/25] Current/Best: 14.31/ 20.47 GFLOPS | Progress: (10/10) | 9.32 s Done.
-
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 19/25] Current/Best: 9.19/ 19.04 GFLOPS | Progress: (4/10) | 5.30 s
[Task 19/25] Current/Best: 18.99/ 19.04 GFLOPS | Progress: (8/10) | 7.15 s
[Task 19/25] Current/Best: 18.66/ 19.04 GFLOPS | Progress: (10/10) | 8.54 s Done.
-
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 20/25] Current/Best: 3.12/ 17.86 GFLOPS | Progress: (4/10) | 2.73 s
[Task 20/25] Current/Best: 20.73/ 20.73 GFLOPS | Progress: (8/10) | 4.70 s
[Task 20/25] Current/Best: 17.47/ 20.73 GFLOPS | Progress: (10/10) | 6.85 s
[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 21/25] Current/Best: 10.86/ 14.28 GFLOPS | Progress: (4/10) | 2.95 s
[Task 21/25] Current/Best: 13.95/ 20.94 GFLOPS | Progress: (8/10) | 7.91 s
[Task 21/25] Current/Best: 16.36/ 20.94 GFLOPS | Progress: (10/10) | 8.43 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 22/25] Current/Best: 10.48/ 14.86 GFLOPS | Progress: (4/10) | 3.42 s
[Task 22/25] Current/Best: 15.16/ 16.42 GFLOPS | Progress: (8/10) | 6.03 s
[Task 22/25] Current/Best: 8.44/ 16.42 GFLOPS | Progress: (10/10) | 7.37
s Done.
-
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 23/25] Current/Best: 12.10/ 14.00 GFLOPS | Progress: (4/10) | 3.77 s
[Task 23/25] Current/Best: 19.71/ 19.71 GFLOPS | Progress: (8/10) | 7.02 s
[Task 23/25] Current/Best: 10.81/ 19.71 GFLOPS | Progress: (10/10) | 7.99 s Done.
-
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 24/25] Current/Best: 4.07/ 4.07 GFLOPS | Progress: (4/10) | 5.50 s
[Task 24/25] Current/Best: 3.38/ 8.30 GFLOPS | Progress: (8/10) | 18.01 s
[Task 24/25] Current/Best: 0.00/ 8.30 GFLOPS | Progress: (10/10) | 228.79 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s Done.
+
[Task 21/25] Current/Best: 7.00/ 17.83 GFLOPS | Progress: (4/10) | 3.10 s
[Task 21/25] Current/Best: 8.89/ 23.44 GFLOPS | Progress: (8/10) | 4.91 s
[Task 21/25] Current/Best: 0.00/ 23.44 GFLOPS | Progress: (10/10) | 5.53 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 22/25] Current/Best: 18.58/ 21.71 GFLOPS | Progress: (4/10) | 2.65 s
[Task 22/25] Current/Best: 14.79/ 21.71 GFLOPS | Progress: (8/10) | 4.44 s
[Task 22/25] Current/Best: 5.94/ 21.71 GFLOPS | Progress: (10/10) | 6.47 s Done.
+
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 23/25] Current/Best: 22.80/ 22.80 GFLOPS | Progress: (4/10) | 5.67 s
[Task 23/25] Current/Best: 20.16/ 22.80 GFLOPS | Progress: (8/10) | 8.49 s
[Task 23/25] Current/Best: 7.01/ 22.80 GFLOPS | Progress: (10/10) | 10.75 s Done.
+
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
[Task 24/25] Current/Best: 1.97/ 4.99 GFLOPS | Progress: (4/10) | 12.00 s
[Task 24/25] Current/Best: 8.84/ 10.59 GFLOPS | Progress: (8/10) | 24.16 s
[Task 24/25] Current/Best: 1.15/ 10.59 GFLOPS | Progress: (10/10) | 26.93 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s Done.
Done.
- Done.
-
[Task 25/25] Current/Best: 6.37/ 7.19 GFLOPS | Progress: (4/10) | 4.10 s
[Task 25/25] Current/Best: 1.55/ 9.03 GFLOPS | Progress: (8/10) | 6.59 s
[Task 25/25] Current/Best: 10.12/ 10.12 GFLOPS | Progress: (10/10) | 19.14 s
+
[Task 25/25] Current/Best: 5.10/ 8.37 GFLOPS | Progress: (4/10) | 3.96 s
[Task 25/25] Current/Best: 3.50/ 8.37 GFLOPS | Progress: (8/10) | 12.39 s
[Task 25/25] Current/Best: 1.48/ 9.49 GFLOPS | Progress: (10/10) | 17.37 s Done.
+
The output from this tuning process will look something like this:
@@ -648,8 +649,8 @@ improvement in comparing the optimized model to the unoptimized model.
.. code-block:: none
- optimized: {'mean': 425.2909051199981, 'median': 425.08289284999137, 'std': 0.6725116699840613}
- unoptimized: {'mean': 496.60638313000044, 'median': 496.4213516499967, 'std': 0.8980274498640965}
+ optimized: {'mean': 433.3818005000001, 'median': 433.01172650000126, 'std': 1.2187799667414965}
+ unoptimized: {'mean': 492.6316053099993, 'median': 492.4662148999971, 'std': 0.45000288160307683}
@@ -669,7 +670,7 @@ profiling/benchmarking.
.. rst-class:: sphx-glr-timing
- **Total running time of the script:** ( 10 minutes 3.950 seconds)
+ **Total running time of the script:** ( 6 minutes 51.260 seconds)
.. _sphx_glr_download_tutorial_autotvm_relay_x86.py:
diff --git a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
index 4347fee4c..135fdb4a6 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -235,7 +235,7 @@ device and returns the measured cost. Network overhead is excluded.
.. code-block:: none
- 1.281e-07 secs/op
+ 1.239e-07 secs/op
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index 3453fb1c8..45b1b07d8 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -233,7 +233,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
.. code-block:: none
- [stage(a, placeholder(a, 0x22cf41b0)), stage(b, placeholder(b, 0x2334a8f0)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(mi [...]
+ [stage(a, placeholder(a, 0x207ddba0)), stage(b, placeholder(b, 0x1a5492f0)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(mi [...]
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index 24287ee22..3d3ea21bb 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,17 +5,17 @@
Computation times
=================
-**13:12.953** total execution time for **tutorial** files:
+**09:45.828** total execution time for **tutorial** files:
-- **10:03.950**: :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)
-- **01:07.232**: :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``)
-- **00:58.830**: :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)
-- **00:34.159**: :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)
-- **00:26.513**: :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)
-- **00:01.118**: :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)
-- **00:00.736**: :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)
-- **00:00.218**: :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``)
-- **00:00.053**: :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``)
-- **00:00.049**: :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``)
-- **00:00.047**: :ref:`sphx_glr_tutorial_install.py` (``install.py``)
-- **00:00.047**: :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``)
+- **06:51.260**: :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)
+- **01:03.810**: :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``)
+- **01:00.772**: :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)
+- **00:25.543**: :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)
+- **00:22.333**: :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)
+- **00:01.089**: :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)
+- **00:00.718**: :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)
+- **00:00.176**: :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``)
+- **00:00.034**: :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``)
+- **00:00.032**: :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``)
+- **00:00.031**: :ref:`sphx_glr_tutorial_install.py` (``install.py``)
+- **00:00.029**: :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``)
diff --git a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
index 3715305e7..49dcfa436 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -244,7 +244,7 @@ helper function to run a profile of the TVM generated code.
.. code-block:: none
Numpy running time: 0.000008
- naive: 0.000008
+ naive: 0.000006
@@ -335,7 +335,7 @@ compile and run this new schedule with the parallel operation applied:
.. code-block:: none
- parallel: 0.000006
+ parallel: 0.000007
@@ -388,7 +388,7 @@ factor to be the number of threads on your CPU.
.. code-block:: none
- vector: 0.000025
+ vector: 0.000026
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [(stride: int32*n: int32)], [], type="auto"),
@@ -438,10 +438,10 @@ We can now compare the different schedules
.. code-block:: none
Operator Timing Performance
- numpy 7.732530000339467e-06 1.0
- naive 7.839e-06 1.0137691026941842
- parallel 6.0116e-06 0.7774428291563155
- vector 2.45931e-05 3.1804726265427146
+ numpy 8.116219999010354e-06 1.0
+ naive 5.8296e-06 0.7182653994976514
+ parallel 7.0317e-06 0.866376219577267
+ vector 2.5753399999999996e-05 3.173078108176001
@@ -830,7 +830,7 @@ matrix multiplication.
.. code-block:: none
- Numpy running time: 0.019003
+ Numpy running time: 0.017579
@@ -886,7 +886,7 @@ optimizations.
.. code-block:: none
- none: 3.211525
+ none: 3.421662
@@ -985,7 +985,7 @@ schedule.
.. code-block:: none
- blocking: 0.325972
+ blocking: 0.299119
@@ -1077,7 +1077,7 @@ already cache friendly from our previous optimizations.
.. code-block:: none
- vectorization: 0.350958
+ vectorization: 0.335124
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1149,7 +1149,7 @@ more cache friendly.
.. code-block:: none
- loop permutation: 0.118166
+ loop permutation: 0.112224
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1246,7 +1246,7 @@ optimized schedule.
.. code-block:: none
- array packing: 0.109252
+ array packing: 0.107586
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1337,7 +1337,7 @@ to `C` when all the block results are ready.
.. code-block:: none
- block caching: 0.110593
+ block caching: 0.110072
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1421,7 +1421,7 @@ of thread-level parallelization.
.. code-block:: none
- parallelization: 0.145116
+ parallelization: 0.143778
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1500,13 +1500,13 @@ working, we can compare the results.
.. code-block:: none
Operator Timing Performance
- none 3.2115246612999995 1.0
- blocking 0.325971564 0.10150056386864216
- vectorization 0.3509578926 0.10928077147566884
- loop permutation 0.1181656727 0.03679425978692858
- array packing 0.10925178390000001 0.034018665718660794
- block caching 0.11059276209999999 0.03443621761111837
- parallelization 0.1451155605 0.04518587767632411
+ none 3.4216618965000003 1.0
+ blocking 0.2991194263 0.08741934046901818
+ vectorization 0.3351239071 0.09794185318040817
+ loop permutation 0.1122239282 0.0327980763718336
+ array packing 0.1075860693 0.03144263593374004
+ block caching 0.11007208180000001 0.03216918711711177
+ parallelization 0.143777803 0.04201987436194954
@@ -1541,6 +1541,11 @@ operations with tunable parameters that allows you to automatically optimize
the computation for specific platforms.
+.. rst-class:: sphx-glr-timing
+
+ **Total running time of the script:** ( 1 minutes 0.772 seconds)
+
+
.. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
diff --git a/docs/commit_hash b/docs/commit_hash
index 5d1c57ff4..6efa18321 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-8691cbed0bf32070b931a4341bded7fa0f639826
+bce57586bd3e41ea3c38a157c126f1fea40a8313
diff --git a/docs/how_to/compile_models/from_coreml.html b/docs/how_to/compile_models/from_coreml.html
index c9d93d332..b268ebed3 100644
--- a/docs/how_to/compile_models/from_coreml.html
+++ b/docs/how_to/compile_models/from_coreml.html
@@ -220,6 +220,7 @@
</li>
<li class="toctree-l3"><a class="reference internal" href="from_darknet.html">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_paddle.html">Compile PaddlePaddle Models</a></li>
+<li class="toctree-l3"><a class="reference internal" href="from_oneflow.html">Compile OneFlow Models</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../deploy/index.html">Deploy Models and Integrate TVM</a></li>
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index af92d3ae1..c1cf81d8a 100644
--- a/docs/how_to/compile_models/from_darknet.html
+++ b/docs/how_to/compile_models/from_darknet.html
@@ -220,6 +220,7 @@
</ul>
</li>
<li class="toctree-l3"><a class="reference internal" href="from_paddle.html">Compile PaddlePaddle Models</a></li>
+<li class="toctree-l3"><a class="reference internal" href="from_oneflow.html">Compile OneFlow Models</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../deploy/index.html">Deploy Models and Integrate TVM</a></li>
diff --git a/docs/how_to/compile_models/from_keras.html b/docs/how_to/compile_models/from_keras.html
index 29dab2989..5ce10b852 100644
--- a/docs/how_to/compile_models/from_keras.html
+++ b/docs/how_to/compile_models/from_keras.html
@@ -220,6 +220,7 @@
<li class="toctree-l3"><a class="reference internal" href="from_coreml.html">Compile CoreML Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_darknet.html">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_paddle.html">Compile PaddlePaddle Models</a></li>
+<li class="toctree-l3"><a class="reference internal" href="from_oneflow.html">Compile OneFlow Models</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../deploy/index.html">Deploy Models and Integrate TVM</a></li>
diff --git a/docs/how_to/compile_models/from_mxnet.html b/docs/how_to/compile_models/from_mxnet.html
index f39c97e8d..113fe41f9 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -219,6 +219,7 @@
<li class="toctree-l3"><a class="reference internal" href="from_coreml.html">Compile CoreML Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_darknet.html">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_paddle.html">Compile PaddlePaddle Models</a></li>
+<li class="toctree-l3"><a class="reference internal" href="from_oneflow.html">Compile OneFlow Models</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../deploy/index.html">Deploy Models and Integrate TVM</a></li>
@@ -400,7 +401,7 @@
</div>
<img alt="../../_images/sphx_glr_from_mxnet_001.png" class="sphx-glr-single-img" src="../../_images/sphx_glr_from_mxnet_001.png" />
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip33034771-7792-4214-abe4-5ba2d8d9fbb2 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip67880911-6b49-4538-938a-4b8f125626ab from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
x (1, 3, 224, 224)
</pre></div>
</div>
diff --git a/docs/how_to/compile_models/from_pytorch.html b/docs/how_to/compile_models/from_oneflow.html
similarity index 62%
copy from docs/how_to/compile_models/from_pytorch.html
copy to docs/how_to/compile_models/from_oneflow.html
index 73ac20076..fdb24feca 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -11,7 +11,7 @@
<meta name="viewport" content="width=device-width, initial-scale=1.0">
- <title>Compile PyTorch Models — tvm 0.9.dev0 documentation</title>
+ <title>Compile OneFlow Models — tvm 0.9.dev0 documentation</title>
@@ -45,8 +45,8 @@
<script type="text/javascript" src="../../_static/js/tlcpack_theme.js"></script>
<link rel="index" title="Index" href="../../genindex.html" />
<link rel="search" title="Search" href="../../search.html" />
- <link rel="next" title="Compile Tensorflow Models" href="from_tensorflow.html" />
- <link rel="prev" title="Compile Deep Learning Models" href="index.html" />
+ <link rel="next" title="Deploy Models and Integrate TVM" href="../deploy/index.html" />
+ <link rel="prev" title="Compile PaddlePaddle Models" href="from_paddle.html" />
</head>
<body class="wy-body-for-nav">
@@ -204,15 +204,7 @@
<li class="toctree-l1"><a class="reference internal" href="../../tutorial/index.html">User Tutorial</a></li>
<li class="toctree-l1 current"><a class="reference internal" href="../index.html">How To Guides</a><ul class="current">
<li class="toctree-l2 current"><a class="reference internal" href="index.html">Compile Deep Learning Models</a><ul class="current">
-<li class="toctree-l3 current"><a class="current reference internal" href="#">Compile PyTorch Models</a><ul>
-<li class="toctree-l4"><a class="reference internal" href="#load-a-pretrained-pytorch-model">Load a pretrained PyTorch model</a></li>
-<li class="toctree-l4"><a class="reference internal" href="#load-a-test-image">Load a test image</a></li>
-<li class="toctree-l4"><a class="reference internal" href="#import-the-graph-to-relay">Import the graph to Relay</a></li>
-<li class="toctree-l4"><a class="reference internal" href="#relay-build">Relay Build</a></li>
-<li class="toctree-l4"><a class="reference internal" href="#execute-the-portable-graph-on-tvm">Execute the portable graph on TVM</a></li>
-<li class="toctree-l4"><a class="reference internal" href="#look-up-synset-name">Look up synset name</a></li>
-</ul>
-</li>
+<li class="toctree-l3"><a class="reference internal" href="from_pytorch.html">Compile PyTorch Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_tensorflow.html">Compile Tensorflow Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_mxnet.html">Compile MXNet Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_onnx.html">Compile ONNX Models</a></li>
@@ -221,6 +213,15 @@
<li class="toctree-l3"><a class="reference internal" href="from_coreml.html">Compile CoreML Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_darknet.html">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_paddle.html">Compile PaddlePaddle Models</a></li>
+<li class="toctree-l3 current"><a class="current reference internal" href="#">Compile OneFlow Models</a><ul>
+<li class="toctree-l4"><a class="reference internal" href="#load-a-pretrained-oneflow-model-and-save-model">Load a pretrained OneFlow model and save model</a></li>
+<li class="toctree-l4"><a class="reference internal" href="#load-a-test-image">Load a test image</a></li>
+<li class="toctree-l4"><a class="reference internal" href="#import-the-graph-to-relay">Import the graph to Relay</a></li>
+<li class="toctree-l4"><a class="reference internal" href="#relay-build">Relay Build</a></li>
+<li class="toctree-l4"><a class="reference internal" href="#execute-the-portable-graph-on-tvm">Execute the portable graph on TVM</a></li>
+<li class="toctree-l4"><a class="reference internal" href="#look-up-synset-name">Look up synset name</a></li>
+</ul>
+</li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../deploy/index.html">Deploy Models and Integrate TVM</a></li>
@@ -318,13 +319,13 @@
<li><a href="index.html">Compile Deep Learning Models</a> <span class="br-arrow">></span></li>
- <li>Compile PyTorch Models</li>
+ <li>Compile OneFlow Models</li>
<li class="wy-breadcrumbs-aside">
- <a href="../../_sources/how_to/compile_models/from_pytorch.rst.txt" rel="nofollow"> <img src="../../_static//img/source.svg" alt="viewsource"/></a>
+ <a href="../../_sources/how_to/compile_models/from_oneflow.rst.txt" rel="nofollow"> <img src="../../_static//img/source.svg" alt="viewsource"/></a>
</li>
@@ -339,55 +340,167 @@
<div class="sphx-glr-download-link-note admonition note">
<p class="admonition-title">Note</p>
-<p>Click <a class="reference internal" href="#sphx-glr-download-how-to-compile-models-from-pytorch-py"><span class="std std-ref">here</span></a> to download the full example code</p>
+<p>Click <a class="reference internal" href="#sphx-glr-download-how-to-compile-models-from-oneflow-py"><span class="std std-ref">here</span></a> to download the full example code</p>
</div>
-<div class="sphx-glr-example-title section" id="compile-pytorch-models">
-<span id="sphx-glr-how-to-compile-models-from-pytorch-py"></span><h1>Compile PyTorch Models<a class="headerlink" href="#compile-pytorch-models" title="Permalink to this headline">¶</a></h1>
-<p><strong>Author</strong>: <a class="reference external" href="https://github.com/alexwong/">Alex Wong</a></p>
-<p>This article is an introductory tutorial to deploy PyTorch models with Relay.</p>
-<p>For us to begin with, PyTorch should be installed.
-TorchVision is also required since we will be using it as our model zoo.</p>
+<div class="sphx-glr-example-title section" id="compile-oneflow-models">
+<span id="sphx-glr-how-to-compile-models-from-oneflow-py"></span><h1>Compile OneFlow Models<a class="headerlink" href="#compile-oneflow-models" title="Permalink to this headline">¶</a></h1>
+<p><strong>Author</strong>: <a class="reference external" href="https://github.com/BBuf/">Xiaoyu Zhang</a></p>
+<p>This article is an introductory tutorial to deploy OneFlow models with Relay.</p>
+<p>For us to begin with, OneFlow package should be installed.</p>
<p>A quick solution is to install via pip</p>
-<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>pip install <span class="nv">torch</span><span class="o">==</span><span class="m">1</span>.7.0
-pip install <span class="nv">torchvision</span><span class="o">==</span><span class="m">0</span>.8.1
+<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>pip install <span class="nv">flowvision</span><span class="o">==</span><span class="m">0</span>.1.0
+python3 -m pip install -f https://release.oneflow.info <span class="nv">oneflow</span><span class="o">==</span><span class="m">0</span>.7.0+cpu
</pre></div>
</div>
-<p>or please refer to official site
-<a class="reference external" href="https://pytorch.org/get-started/locally/">https://pytorch.org/get-started/locally/</a></p>
-<p>PyTorch versions should be backwards compatible but should be used
-with the proper TorchVision version.</p>
-<p>Currently, TVM supports PyTorch 1.7 and 1.4. Other versions may
-be unstable.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">tvm</span>
-<span class="kn">from</span> <span class="nn">tvm</span> <span class="kn">import</span> <span class="n">relay</span>
-
+<p>or please refer to official site:
+<a class="reference external" href="https://github.com/Oneflow-Inc/oneflow">https://github.com/Oneflow-Inc/oneflow</a></p>
+<p>Currently, TVM supports OneFlow 0.7.0. Other versions may be unstable.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">os</span><span class="o">,</span> <span class="nn">math</span>
+<span class="kn">from</span> <span class="nn">matplotlib</span> <span class="kn">import</span> <span class="n">pyplot</span> <span class="k">as</span> <span class="n">plt</span>
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">from</span> <span class="nn">PIL</span> <span class="kn">import</span> <span class="n">Image</span>
-<span class="kn">from</span> <span class="nn">tvm.contrib.download</span> <span class="kn">import</span> <span class="n">download_testdata</span>
+<span class="c1"># oneflow imports</span>
+<span class="kn">import</span> <span class="nn">flowvision</span>
+<span class="kn">import</span> <span class="nn">oneflow</span> <span class="k">as</span> <span class="nn">flow</span>
+<span class="kn">import</span> <span class="nn">oneflow.nn</span> <span class="k">as</span> <span class="nn">nn</span>
-<span class="c1"># PyTorch imports</span>
-<span class="kn">import</span> <span class="nn">torch</span>
-<span class="kn">import</span> <span class="nn">torchvision</span>
+<span class="kn">import</span> <span class="nn">tvm</span>
+<span class="kn">from</span> <span class="nn">tvm</span> <span class="kn">import</span> <span class="n">relay</span>
+<span class="kn">from</span> <span class="nn">tvm.contrib.download</span> <span class="kn">import</span> <span class="n">download_testdata</span>
+</pre></div>
+</div>
+<p class="sphx-glr-script-out">Out:</p>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional_pil.py:193: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
+ def resize(img, size, interpolation=Image.BILINEAR):
+/usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional.py:65: DeprecationWarning: NEAREST is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.NEAREST or Dither.NONE instead.
+ Image.NEAREST: "nearest",
+/usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional.py:66: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
+ Image.BILINEAR: "bilinear",
+/usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional.py:67: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.
+ Image.BICUBIC: "bicubic",
+/usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional.py:68: DeprecationWarning: BOX is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BOX instead.
+ Image.BOX: "box",
+/usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional.py:69: DeprecationWarning: HAMMING is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.HAMMING instead.
+ Image.HAMMING: "hamming",
+/usr/local/lib/python3.7/dist-packages/flowvision/transforms/functional.py:70: DeprecationWarning: LANCZOS is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.LANCZOS instead.
+ Image.LANCZOS: "lanczos",
+/usr/local/lib/python3.7/dist-packages/flowvision/data/auto_augment.py:28: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
+ _RANDOM_INTERPOLATION = (Image.BILINEAR, Image.BICUBIC)
+/usr/local/lib/python3.7/dist-packages/flowvision/data/auto_augment.py:28: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.
+ _RANDOM_INTERPOLATION = (Image.BILINEAR, Image.BICUBIC)
</pre></div>
</div>
-<div class="section" id="load-a-pretrained-pytorch-model">
-<h2>Load a pretrained PyTorch model<a class="headerlink" href="#load-a-pretrained-pytorch-model" title="Permalink to this headline">¶</a></h2>
+<div class="section" id="load-a-pretrained-oneflow-model-and-save-model">
+<h2>Load a pretrained OneFlow model and save model<a class="headerlink" href="#load-a-pretrained-oneflow-model-and-save-model" title="Permalink to this headline">¶</a></h2>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">model_name</span> <span class="o">=</span> <span class="s2">"resnet18"</span>
-<span class="n">model</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">torchvision</span><span class="o">.</span><span class="n">models</span><span class="p">,</span> <span class="n">model_name</span><span class="p">)(</span><span class="n">pretrained</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+<span class="n">model</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">flowvision</span><span class="o">.</span><span class="n">models</span><span class="p">,</span> <span class="n">model_name</span><span class="p">)(</span><span class="n">pretrained</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
<span class="n">model</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
-<span class="c1"># We grab the TorchScripted model via tracing</span>
-<span class="n">input_shape</span> <span class="o">=</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">]</span>
-<span class="n">input_data</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">randn</span><span class="p">(</span><span class="n">input_shape</span><span class="p">)</span>
-<span class="n">scripted_model</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">trace</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">input_data</span><span class="p">)</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+<span class="n">model_dir</span> <span class="o">=</span> <span class="s2">"resnet18_model"</span>
+<span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">model_dir</span><span class="p">):</span>
+ <span class="n">flow</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="n">model</span><span class="o">.</span><span class="n">state_dict</span><span class="p">(),</span> <span class="n">model_dir</span><span class="p">)</span>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
-
- 0%| | 0.00/44.7M [00:00<?, ?B/s]
- 43%|####3 | 19.3M/44.7M [00:00<00:00, 203MB/s]
-100%|##########| 44.7M/44.7M [00:00<00:00, 235MB/s]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
+
+ 0%| | 0.00/41.5M [00:00<?, ?B/s]
+ 0%| | 16.0k/41.5M [00:00<08:23, 86.3kB/s]
+ 0%| | 48.0k/41.5M [00:00<05:20, 135kB/s]
+ 0%| | 104k/41.5M [00:00<03:28, 208kB/s]
+ 0%| | 208k/41.5M [00:00<02:07, 339kB/s]
+ 1%| | 288k/41.5M [00:00<01:38, 437kB/s]
+ 1%| | 416k/41.5M [00:00<01:08, 633kB/s]
+ 1%|1 | 528k/41.5M [00:01<00:57, 744kB/s]
+ 2%|1 | 672k/41.5M [00:01<00:46, 911kB/s]
+ 2%|1 | 792k/41.5M [00:01<00:44, 957kB/s]
+ 2%|2 | 936k/41.5M [00:01<00:40, 1.06MB/s]
+ 2%|2 | 1.03M/41.5M [00:01<00:40, 1.06MB/s]
+ 3%|2 | 1.19M/41.5M [00:01<00:35, 1.19MB/s]
+ 3%|3 | 1.31M/41.5M [00:01<00:35, 1.17MB/s]
+ 4%|3 | 1.48M/41.5M [00:01<00:31, 1.31MB/s]
+ 4%|3 | 1.61M/41.5M [00:01<00:32, 1.28MB/s]
+ 5%|4 | 1.91M/41.5M [00:02<00:30, 1.38MB/s]
+ 5%|5 | 2.23M/41.5M [00:02<00:24, 1.70MB/s]
+ 6%|5 | 2.41M/41.5M [00:02<00:23, 1.72MB/s]
+ 6%|6 | 2.58M/41.5M [00:02<00:27, 1.47MB/s]
+ 7%|6 | 2.77M/41.5M [00:02<00:25, 1.57MB/s]
+ 7%|7 | 2.92M/41.5M [00:02<00:26, 1.52MB/s]
+ 8%|7 | 3.12M/41.5M [00:02<00:24, 1.66MB/s]
+ 8%|7 | 3.29M/41.5M [00:03<00:24, 1.64MB/s]
+ 9%|8 | 3.68M/41.5M [00:03<00:21, 1.81MB/s]
+ 10%|9 | 4.09M/41.5M [00:03<00:20, 1.92MB/s]
+ 11%|# | 4.52M/41.5M [00:03<00:19, 2.03MB/s]
+ 12%|#1 | 4.96M/41.5M [00:03<00:16, 2.35MB/s]
+ 13%|#2 | 5.23M/41.5M [00:03<00:17, 2.23MB/s]
+ 13%|#3 | 5.45M/41.5M [00:04<00:17, 2.12MB/s]
+ 14%|#4 | 5.92M/41.5M [00:04<00:14, 2.50MB/s]
+ 15%|#4 | 6.21M/41.5M [00:04<00:14, 2.60MB/s]
+ 16%|#5 | 6.47M/41.5M [00:04<00:16, 2.26MB/s]
+ 16%|#6 | 6.74M/41.5M [00:04<00:15, 2.39MB/s]
+ 17%|#6 | 6.98M/41.5M [00:04<00:15, 2.33MB/s]
+ 18%|#7 | 7.30M/41.5M [00:04<00:14, 2.54MB/s]
+ 18%|#8 | 7.55M/41.5M [00:04<00:14, 2.47MB/s]
+ 19%|#8 | 7.88M/41.5M [00:05<00:13, 2.70MB/s]
+ 20%|#9 | 8.14M/41.5M [00:05<00:13, 2.67MB/s]
+ 20%|## | 8.46M/41.5M [00:05<00:12, 2.86MB/s]
+ 21%|##1 | 8.74M/41.5M [00:05<00:12, 2.79MB/s]
+ 22%|##1 | 9.09M/41.5M [00:05<00:11, 2.93MB/s]
+ 23%|##2 | 9.37M/41.5M [00:05<00:12, 2.80MB/s]
+ 23%|##3 | 9.73M/41.5M [00:05<00:10, 3.05MB/s]
+ 24%|##4 | 10.0M/41.5M [00:05<00:11, 2.96MB/s]
+ 25%|##5 | 10.4M/41.5M [00:05<00:10, 3.23MB/s]
+ 26%|##5 | 10.7M/41.5M [00:06<00:10, 3.17MB/s]
+ 27%|##6 | 11.1M/41.5M [00:06<00:09, 3.38MB/s]
+ 28%|##7 | 11.5M/41.5M [00:06<00:09, 3.32MB/s]
+ 29%|##8 | 11.9M/41.5M [00:06<00:08, 3.56MB/s]
+ 29%|##9 | 12.2M/41.5M [00:06<00:08, 3.55MB/s]
+ 31%|### | 12.7M/41.5M [00:06<00:08, 3.72MB/s]
+ 31%|###1 | 13.0M/41.5M [00:06<00:08, 3.66MB/s]
+ 32%|###2 | 13.5M/41.5M [00:06<00:07, 3.85MB/s]
+ 33%|###3 | 13.8M/41.5M [00:06<00:07, 3.75MB/s]
+ 34%|###4 | 14.3M/41.5M [00:06<00:07, 4.00MB/s]
+ 35%|###5 | 14.7M/41.5M [00:07<00:07, 3.88MB/s]
+ 37%|###6 | 15.2M/41.5M [00:07<00:06, 4.22MB/s]
+ 38%|###7 | 15.6M/41.5M [00:07<00:06, 4.14MB/s]
+ 39%|###8 | 16.1M/41.5M [00:07<00:05, 4.51MB/s]
+ 40%|###9 | 16.6M/41.5M [00:07<00:05, 4.38MB/s]
+ 41%|####1 | 17.1M/41.5M [00:07<00:05, 4.69MB/s]
+ 42%|####2 | 17.6M/41.5M [00:07<00:05, 4.58MB/s]
+ 44%|####3 | 18.1M/41.5M [00:07<00:05, 4.80MB/s]
+ 45%|####4 | 18.6M/41.5M [00:07<00:05, 4.58MB/s]
+ 46%|####6 | 19.2M/41.5M [00:08<00:04, 4.97MB/s]
+ 47%|####7 | 19.7M/41.5M [00:08<00:04, 4.96MB/s]
+ 49%|####8 | 20.3M/41.5M [00:08<00:04, 5.38MB/s]
+ 50%|##### | 20.8M/41.5M [00:08<00:04, 5.38MB/s]
+ 52%|#####1 | 21.5M/41.5M [00:08<00:03, 5.81MB/s]
+ 53%|#####3 | 22.1M/41.5M [00:08<00:03, 5.79MB/s]
+ 55%|#####4 | 22.7M/41.5M [00:08<00:03, 6.08MB/s]
+ 56%|#####6 | 23.3M/41.5M [00:08<00:03, 6.05MB/s]
+ 58%|#####7 | 24.0M/41.5M [00:08<00:02, 6.29MB/s]
+ 59%|#####9 | 24.6M/41.5M [00:09<00:02, 6.26MB/s]
+ 61%|######1 | 25.3M/41.5M [00:09<00:02, 6.55MB/s]
+ 63%|######2 | 26.0M/41.5M [00:09<00:02, 6.51MB/s]
+ 64%|######4 | 26.8M/41.5M [00:09<00:02, 6.90MB/s]
+ 66%|######6 | 27.4M/41.5M [00:09<00:02, 6.79MB/s]
+ 68%|######8 | 28.2M/41.5M [00:09<00:01, 7.25MB/s]
+ 70%|######9 | 28.9M/41.5M [00:09<00:01, 7.14MB/s]
+ 72%|#######2 | 29.9M/41.5M [00:09<00:01, 8.09MB/s]
+ 74%|#######3 | 30.7M/41.5M [00:09<00:01, 8.05MB/s]
+ 76%|#######5 | 31.5M/41.5M [00:09<00:01, 8.18MB/s]
+ 78%|#######7 | 32.3M/41.5M [00:10<00:01, 7.88MB/s]
+ 80%|#######9 | 33.1M/41.5M [00:10<00:01, 7.88MB/s]
+ 81%|########1 | 33.8M/41.5M [00:10<00:01, 7.53MB/s]
+ 84%|########3 | 34.7M/41.5M [00:10<00:00, 7.60MB/s]
+ 86%|########5 | 35.5M/41.5M [00:10<00:00, 7.87MB/s]
+ 87%|########7 | 36.3M/41.5M [00:10<00:00, 6.91MB/s]
+ 91%|######### | 37.6M/41.5M [00:10<00:00, 8.30MB/s]
+ 93%|#########2| 38.4M/41.5M [00:10<00:00, 8.34MB/s]
+ 95%|#########4| 39.2M/41.5M [00:11<00:00, 6.81MB/s]
+ 97%|#########7| 40.4M/41.5M [00:11<00:00, 8.06MB/s]
+ 99%|#########9| 41.2M/41.5M [00:11<00:00, 8.05MB/s]
+100%|##########| 41.5M/41.5M [00:11<00:00, 3.84MB/s]
</pre></div>
</div>
</div>
@@ -401,7 +514,7 @@ be unstable.</p>
<span class="n">img</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span><span class="n">img_path</span><span class="p">)</span><span class="o">.</span><span class="n">resize</span><span class="p">((</span><span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">))</span>
<span class="c1"># Preprocess the image and convert to tensor</span>
-<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">transforms</span>
+<span class="kn">from</span> <span class="nn">flowvision</span> <span class="kn">import</span> <span class="n">transforms</span>
<span class="n">my_preprocess</span> <span class="o">=</span> <span class="n">transforms</span><span class="o">.</span><span class="n">Compose</span><span class="p">(</span>
<span class="p">[</span>
@@ -412,16 +525,27 @@ be unstable.</p>
<span class="p">]</span>
<span class="p">)</span>
<span class="n">img</span> <span class="o">=</span> <span class="n">my_preprocess</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
-<span class="n">img</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">expand_dims</span><span class="p">(</span><span class="n">img</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+<span class="n">img</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">expand_dims</span><span class="p">(</span><span class="n">img</span><span class="o">.</span><span class="n">numpy</span><span class="p">(),</span> <span class="mi">0</span><span class="p">)</span>
</pre></div>
</div>
</div>
<div class="section" id="import-the-graph-to-relay">
<h2>Import the graph to Relay<a class="headerlink" href="#import-the-graph-to-relay" title="Permalink to this headline">¶</a></h2>
-<p>Convert PyTorch graph to Relay graph. The input name can be arbitrary.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">input_name</span> <span class="o">=</span> <span class="s2">"input0"</span>
-<span class="n">shape_list</span> <span class="o">=</span> <span class="p">[(</span><span class="n">input_name</span><span class="p">,</span> <span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">)]</span>
-<span class="n">mod</span><span class="p">,</span> <span class="n">params</span> <span class="o">=</span> <a href="../../reference/api/python/relay/frontend.html#tvm.relay.frontend.from_pytorch" title="View documentation for tvm.relay.frontend.from_pytorch"><span class="n">relay</span><span class="o">.</span><span class="n">frontend</span><span class="o">.</span><span class="n">from_pytorch</span></a><span class="p">(</span><span class="n">scripted_model</span><span class="p">,</span> <s [...]
+<p>Convert OneFlow graph to Relay graph. The input name can be arbitrary.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">class</span> <span class="nc">Graph</span><span class="p">(</span><span class="n">flow</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Graph</span><span class="p">):</span>
+ <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">module</span><span class="p">):</span>
+ <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+ <span class="bp">self</span><span class="o">.</span><span class="n">m</span> <span class="o">=</span> <span class="n">module</span>
+
+ <span class="k">def</span> <span class="nf">build</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">):</span>
+ <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">m</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+ <span class="k">return</span> <span class="n">out</span>
+
+
+<span class="n">graph</span> <span class="o">=</span> <span class="n">Graph</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
+<span class="n">_</span> <span class="o">=</span> <span class="n">graph</span><span class="o">.</span><span class="n">_compile</span><span class="p">(</span><span class="n">flow</span><span class="o">.</span><span class="n">randn</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">))</span>
+
+<span class="n">mod</span><span class="p">,</span> <span class="n">params</span> <span class="o">=</span> <a href="../../reference/api/python/relay/frontend.html#tvm.relay.frontend.from_oneflow" title="View documentation for tvm.relay.frontend.from_oneflow"><span class="n">relay</span><span class="o">.</span><span class="n">frontend</span><span class="o">.</span><span class="n">from_oneflow</span></a><span class="p">(</span><span class="n">graph</span><span class="p">,</span> <span class [...]
</pre></div>
</div>
</div>
@@ -438,16 +562,18 @@ be unstable.</p>
<div class="section" id="execute-the-portable-graph-on-tvm">
<h2>Execute the portable graph on TVM<a class="headerlink" href="#execute-the-portable-graph-on-tvm" title="Permalink to this headline">¶</a></h2>
<p>Now we can try deploying the compiled model on target.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">tvm.contrib</span> <span class="kn">import</span> <span class="n">graph_executor</span>
-
-<span class="n">dtype</span> <span class="o">=</span> <span class="s2">"float32"</span>
-<span class="n">m</span> <span class="o">=</span> <a href="../../reference/api/python/graph_executor.html#tvm.contrib.graph_executor.GraphModule" title="View documentation for tvm.contrib.graph_executor.GraphModule"><span class="n">graph_executor</span><span class="o">.</span><span class="n">GraphModule</span></a><span class="p">(</span><span class="n">lib</span><span class="p">[</span><span class="s2">"default"</span><span class="p">](</span><span class="n">dev</span><span cla [...]
-<span class="c1"># Set inputs</span>
-<span class="n">m</span><span class="o">.</span><span class="n">set_input</span><span class="p">(</span><span class="n">input_name</span><span class="p">,</span> <a href="../../reference/api/python/ndarray.html#tvm.nd.array" title="View documentation for tvm.nd.array"><span class="n">tvm</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">array</span></a><span class="p">(</span><span class="n">img</span><span class="o">.</span><span class="n">as [...]
-<span class="c1"># Execute</span>
-<span class="n">m</span><span class="o">.</span><span class="n">run</span><span class="p">()</span>
-<span class="c1"># Get outputs</span>
-<span class="n">tvm_output</span> <span class="o">=</span> <span class="n">m</span><span class="o">.</span><span class="n">get_output</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">target</span> <span class="o">=</span> <span class="s2">"cuda"</span>
+<span class="k">with</span> <a href="../../reference/api/python/ir.html#tvm.transform.PassContext" title="View documentation for tvm.transform.PassContext"><span class="n">tvm</span><span class="o">.</span><span class="n">transform</span><span class="o">.</span><span class="n">PassContext</span></a><span class="p">(</span><span class="n">opt_level</span><span class="o">=</span><span class="mi">10</span><span class="p">):</span>
+ <span class="n">intrp</span> <span class="o">=</span> <span class="n">relay</span><span class="o">.</span><span class="n">build_module</span><span class="o">.</span><span class="n">create_executor</span><span class="p">(</span><span class="s2">"graph"</span><span class="p">,</span> <span class="n">mod</span><span class="p">,</span> <span class="n">tvm</span><span class="o">.</span><span class="n">cuda</span><span class="p">(</span><span class="mi">0</span><span class="p">), [...]
+
+<span class="nb">print</span><span class="p">(</span><span class="nb">type</span><span class="p">(</span><span class="n">img</span><span class="p">))</span>
+<span class="nb">print</span><span class="p">(</span><span class="n">img</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span>
+<span class="n">tvm_output</span> <span class="o">=</span> <span class="n">intrp</span><span class="o">.</span><span class="n">evaluate</span><span class="p">()(</span><a href="../../reference/api/python/ndarray.html#tvm.nd.array" title="View documentation for tvm.nd.array"><span class="n">tvm</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">array</span></a><span class="p">(</span><span class="n">img</span><span class="o">.</span><span class= [...]
+</pre></div>
+</div>
+<p class="sphx-glr-script-out">Out:</p>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span><class 'numpy.ndarray'>
+(1, 3, 224, 224)
</pre></div>
</div>
</div>
@@ -488,30 +614,32 @@ be unstable.</p>
<span class="n">top1_tvm</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">tvm_output</span><span class="o">.</span><span class="n">numpy</span><span class="p">()[</span><span class="mi">0</span><span class="p">])</span>
<span class="n">tvm_class_key</span> <span class="o">=</span> <span class="n">class_id_to_key</span><span class="p">[</span><span class="n">top1_tvm</span><span class="p">]</span>
-<span class="c1"># Convert input to PyTorch variable and get PyTorch result for comparison</span>
-<span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
- <span class="n">torch_img</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">from_numpy</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
+<span class="c1"># Convert input to OneFlow variable and get OneFlow result for comparison</span>
+<span class="k">with</span> <span class="n">flow</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+ <span class="n">torch_img</span> <span class="o">=</span> <span class="n">flow</span><span class="o">.</span><span class="n">from_numpy</span><span class="p">(</span><span class="n">img</span><span class="p">)</span>
<span class="n">output</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="n">torch_img</span><span class="p">)</span>
- <span class="c1"># Get top-1 result for PyTorch</span>
- <span class="n">top1_torch</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">numpy</span><span class="p">())</span>
- <span class="n">torch_class_key</span> <span class="o">=</span> <span class="n">class_id_to_key</span><span class="p">[</span><span class="n">top1_torch</span><span class="p">]</span>
+ <span class="c1"># Get top-1 result for OneFlow</span>
+ <span class="n">top_oneflow</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">numpy</span><span class="p">())</span>
+ <span class="n">oneflow_class_key</span> <span class="o">=</span> <span class="n">class_id_to_key</span><span class="p">[</span><span class="n">top_oneflow</span><span class="p">]</span>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Relay top-1 id: </span><span class="si">{}</span><span class="s2">, class name: </span><span class="si">{}</span><span class="s2">"</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">top1_tvm</span><span class="p">,</span> <span class="n">key_to_classname</span><span class="p">[</span><span class="n">tvm_class_key</span><span class="p">]))</span>
-<span class="nb">print</span><span class="p">(</span><span class="s2">"Torch top-1 id: </span><span class="si">{}</span><span class="s2">, class name: </span><span class="si">{}</span><span class="s2">"</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">top1_torch</span><span class="p">,</span> <span class="n">key_to_classname</span><span class="p">[</span><span class="n">torch_class_key</span><span class="p">]))</span>
+<span class="nb">print</span><span class="p">(</span>
+ <span class="s2">"OneFlow top-1 id: </span><span class="si">{}</span><span class="s2">, class name: </span><span class="si">{}</span><span class="s2">"</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">top_oneflow</span><span class="p">,</span> <span class="n">key_to_classname</span><span class="p">[</span><span class="n">oneflow_class_key</span><span class="p">])</span>
+<span class="p">)</span>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Relay top-1 id: 281, class name: tabby, tabby cat
-Torch top-1 id: 281, class name: tabby, tabby cat
+OneFlow top-1 id: 281, class name: tabby, tabby cat
</pre></div>
</div>
-<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-pytorch-py">
+<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-oneflow-py">
<div class="sphx-glr-download docutils container">
-<p><a class="reference download internal" download="" href="../../_downloads/f90d5f6bfd99e0d9812ae5b91503e148/from_pytorch.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_pytorch.py</span></code></a></p>
+<p><a class="reference download internal" download="" href="../../_downloads/f7ae979fbe61064749ce0fb7a621eb4c/from_oneflow.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_oneflow.py</span></code></a></p>
</div>
<div class="sphx-glr-download docutils container">
-<p><a class="reference download internal" download="" href="../../_downloads/1f4943aed1aa607b2775c18b1d71db10/from_pytorch.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">from_pytorch.ipynb</span></code></a></p>
+<p><a class="reference download internal" download="" href="../../_downloads/2e7b51cb39c472626dd3f046d9b89966/from_oneflow.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">from_oneflow.ipynb</span></code></a></p>
</div>
</div>
<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
@@ -528,10 +656,10 @@ Torch top-1 id: 281, class name: tabby, tabby cat
<div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
- <a href="from_tensorflow.html" class="btn btn-neutral float-right" title="Compile Tensorflow Models" accesskey="n" rel="next">Next <span class="fa fa-arrow-circle-right"></span></a>
+ <a href="../deploy/index.html" class="btn btn-neutral float-right" title="Deploy Models and Integrate TVM" accesskey="n" rel="next">Next <span class="fa fa-arrow-circle-right"></span></a>
- <a href="index.html" class="btn btn-neutral float-left" title="Compile Deep Learning Models" accesskey="p" rel="prev"><span class="fa fa-arrow-circle-left"></span> Previous</a>
+ <a href="from_paddle.html" class="btn btn-neutral float-left" title="Compile PaddlePaddle Models" accesskey="p" rel="prev"><span class="fa fa-arrow-circle-left"></span> Previous</a>
</div>
diff --git a/docs/how_to/compile_models/from_onnx.html b/docs/how_to/compile_models/from_onnx.html
index 17c40329e..a1c312f8a 100644
--- a/docs/how_to/compile_models/from_onnx.html
+++ b/docs/how_to/compile_models/from_onnx.html
@@ -221,6 +221,7 @@
<li class="toctree-l3"><a class="reference internal" href="from_coreml.html">Compile CoreML Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_darknet.html">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_paddle.html">Compile PaddlePaddle Models</a></li>
+<li class="toctree-l3"><a class="reference internal" href="from_oneflow.html">Compile OneFlow Models</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../deploy/index.html">Deploy Models and Integrate TVM</a></li>
diff --git a/docs/how_to/compile_models/from_paddle.html b/docs/how_to/compile_models/from_paddle.html
index c27c32dae..3a9c4c30f 100644
--- a/docs/how_to/compile_models/from_paddle.html
+++ b/docs/how_to/compile_models/from_paddle.html
@@ -45,7 +45,7 @@
<script type="text/javascript" src="../../_static/js/tlcpack_theme.js"></script>
<link rel="index" title="Index" href="../../genindex.html" />
<link rel="search" title="Search" href="../../search.html" />
- <link rel="next" title="Deploy Models and Integrate TVM" href="../deploy/index.html" />
+ <link rel="next" title="Compile OneFlow Models" href="from_oneflow.html" />
<link rel="prev" title="Compile YOLO-V2 and YOLO-V3 in DarkNet Models" href="from_darknet.html" />
</head>
@@ -220,6 +220,7 @@
<li class="toctree-l4"><a class="reference internal" href="#look-up-synset-name">Look up synset name</a></li>
</ul>
</li>
+<li class="toctree-l3"><a class="reference internal" href="from_oneflow.html">Compile OneFlow Models</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../deploy/index.html">Deploy Models and Integrate TVM</a></li>
@@ -463,7 +464,7 @@ A quick solution is</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>TVM prediction top-1 id: 282, class name: 282: 'tiger cat',
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 9.448 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 11.807 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-paddle-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/16269b77359771348d507395692524cf/from_paddle.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_paddle.py</span></code></a></p>
@@ -486,7 +487,7 @@ A quick solution is</p>
<div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
- <a href="../deploy/index.html" class="btn btn-neutral float-right" title="Deploy Models and Integrate TVM" accesskey="n" rel="next">Next <span class="fa fa-arrow-circle-right"></span></a>
+ <a href="from_oneflow.html" class="btn btn-neutral float-right" title="Compile OneFlow Models" accesskey="n" rel="next">Next <span class="fa fa-arrow-circle-right"></span></a>
<a href="from_darknet.html" class="btn btn-neutral float-left" title="Compile YOLO-V2 and YOLO-V3 in DarkNet Models" accesskey="p" rel="prev"><span class="fa fa-arrow-circle-left"></span> Previous</a>
diff --git a/docs/how_to/compile_models/from_pytorch.html b/docs/how_to/compile_models/from_pytorch.html
index 73ac20076..89266b8e2 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -221,6 +221,7 @@
<li class="toctree-l3"><a class="reference internal" href="from_coreml.html">Compile CoreML Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_darknet.html">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_paddle.html">Compile PaddlePaddle Models</a></li>
+<li class="toctree-l3"><a class="reference internal" href="from_oneflow.html">Compile OneFlow Models</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../deploy/index.html">Deploy Models and Integrate TVM</a></li>
@@ -386,8 +387,41 @@ be unstable.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
0%| | 0.00/44.7M [00:00<?, ?B/s]
- 43%|####3 | 19.3M/44.7M [00:00<00:00, 203MB/s]
-100%|##########| 44.7M/44.7M [00:00<00:00, 235MB/s]
+ 1%|1 | 512k/44.7M [00:00<00:09, 4.97MB/s]
+ 5%|4 | 2.04M/44.7M [00:00<00:03, 11.4MB/s]
+ 9%|8 | 3.82M/44.7M [00:00<00:02, 14.7MB/s]
+ 13%|#2 | 5.67M/44.7M [00:00<00:02, 16.5MB/s]
+ 16%|#6 | 7.26M/44.7M [00:00<00:02, 14.0MB/s]
+ 19%|#9 | 8.65M/44.7M [00:00<00:03, 11.9MB/s]
+ 22%|##2 | 9.86M/44.7M [00:00<00:03, 10.7MB/s]
+ 24%|##4 | 10.9M/44.7M [00:01<00:03, 10.1MB/s]
+ 27%|##6 | 11.9M/44.7M [00:01<00:03, 10.0MB/s]
+ 29%|##8 | 12.9M/44.7M [00:01<00:03, 9.61MB/s]
+ 32%|###1 | 14.2M/44.7M [00:01<00:02, 10.7MB/s]
+ 36%|###6 | 16.1M/44.7M [00:01<00:02, 13.1MB/s]
+ 40%|#### | 17.9M/44.7M [00:01<00:02, 13.8MB/s]
+ 43%|####3 | 19.2M/44.7M [00:01<00:02, 12.8MB/s]
+ 46%|####6 | 20.7M/44.7M [00:01<00:01, 13.4MB/s]
+ 49%|####9 | 22.0M/44.7M [00:01<00:01, 13.4MB/s]
+ 52%|#####2 | 23.3M/44.7M [00:01<00:01, 13.3MB/s]
+ 55%|#####5 | 24.6M/44.7M [00:02<00:01, 12.1MB/s]
+ 58%|#####7 | 25.8M/44.7M [00:02<00:01, 10.6MB/s]
+ 60%|###### | 26.9M/44.7M [00:02<00:01, 11.0MB/s]
+ 63%|######3 | 28.3M/44.7M [00:02<00:01, 11.8MB/s]
+ 66%|######6 | 29.5M/44.7M [00:02<00:01, 10.8MB/s]
+ 68%|######8 | 30.5M/44.7M [00:02<00:01, 8.73MB/s]
+ 71%|####### | 31.6M/44.7M [00:02<00:01, 9.15MB/s]
+ 73%|#######2 | 32.5M/44.7M [00:03<00:01, 9.05MB/s]
+ 76%|#######5 | 33.8M/44.7M [00:03<00:01, 9.97MB/s]
+ 78%|#######8 | 35.0M/44.7M [00:03<00:00, 10.4MB/s]
+ 81%|######## | 36.0M/44.7M [00:03<00:00, 10.1MB/s]
+ 84%|########3 | 37.5M/44.7M [00:03<00:00, 11.3MB/s]
+ 88%|########8 | 39.3M/44.7M [00:03<00:00, 12.1MB/s]
+ 91%|######### | 40.5M/44.7M [00:03<00:00, 10.3MB/s]
+ 93%|#########3| 41.6M/44.7M [00:03<00:00, 10.5MB/s]
+ 96%|#########5| 42.7M/44.7M [00:04<00:00, 9.42MB/s]
+ 99%|#########8| 44.1M/44.7M [00:04<00:00, 9.00MB/s]
+100%|##########| 44.7M/44.7M [00:04<00:00, 11.0MB/s]
</pre></div>
</div>
</div>
diff --git a/docs/how_to/compile_models/from_tensorflow.html b/docs/how_to/compile_models/from_tensorflow.html
index 4052589ba..0613f676e 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -224,6 +224,7 @@
<li class="toctree-l3"><a class="reference internal" href="from_coreml.html">Compile CoreML Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_darknet.html">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_paddle.html">Compile PaddlePaddle Models</a></li>
+<li class="toctree-l3"><a class="reference internal" href="from_oneflow.html">Compile OneFlow Models</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../deploy/index.html">Deploy Models and Integrate TVM</a></li>
@@ -606,7 +607,7 @@ banana (score = 0.00022)
desk (score = 0.00019)
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 1.729 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 3.238 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-tensorflow-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7f1d3d1b878694c201c614c807cdebc8/from_tensorflow.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_tensorflow.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/from_tflite.html b/docs/how_to/compile_models/from_tflite.html
index 5b5bbe504..311a72aee 100644
--- a/docs/how_to/compile_models/from_tflite.html
+++ b/docs/how_to/compile_models/from_tflite.html
@@ -221,6 +221,7 @@
<li class="toctree-l3"><a class="reference internal" href="from_coreml.html">Compile CoreML Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_darknet.html">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_paddle.html">Compile PaddlePaddle Models</a></li>
+<li class="toctree-l3"><a class="reference internal" href="from_oneflow.html">Compile OneFlow Models</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../deploy/index.html">Deploy Models and Integrate TVM</a></li>
diff --git a/docs/how_to/compile_models/index.html b/docs/how_to/compile_models/index.html
index 781e53781..c6a54799d 100644
--- a/docs/how_to/compile_models/index.html
+++ b/docs/how_to/compile_models/index.html
@@ -213,6 +213,7 @@
<li class="toctree-l3"><a class="reference internal" href="from_coreml.html">Compile CoreML Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_darknet.html">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</a></li>
<li class="toctree-l3"><a class="reference internal" href="from_paddle.html">Compile PaddlePaddle Models</a></li>
+<li class="toctree-l3"><a class="reference internal" href="from_oneflow.html">Compile OneFlow Models</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../deploy/index.html">Deploy Models and Integrate TVM</a></li>
@@ -385,6 +386,12 @@ formats. These how-tos demostrate how to import models using the Python API.</p>
</div>
</div><div class="toctree-wrapper compound">
</div>
+<div class="sphx-glr-thumbcontainer" tooltip="This article is an introductory tutorial to deploy OneFlow models with Relay."><div class="figure align-default" id="id10">
+<img alt="../../_images/sphx_glr_from_oneflow_thumb.png" src="../../_images/sphx_glr_from_oneflow_thumb.png" />
+<p class="caption"><span class="caption-text"><a class="reference internal" href="from_oneflow.html#sphx-glr-how-to-compile-models-from-oneflow-py"><span class="std std-ref">Compile OneFlow Models</span></a></span><a class="headerlink" href="#id10" title="Permalink to this image">¶</a></p>
+</div>
+</div><div class="toctree-wrapper compound">
+</div>
<div style='clear:both'></div><p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
</div>
diff --git a/docs/how_to/compile_models/sg_execution_times.html b/docs/how_to/compile_models/sg_execution_times.html
index 711f3eb0b..a719bd1e6 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -300,17 +300,18 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-compile-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>04:54.129</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>05:32.910</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
<ul class="simple">
-<li><p><strong>01:09.448</strong>: <a class="reference internal" href="from_paddle.html#sphx-glr-how-to-compile-models-from-paddle-py"><span class="std std-ref">Compile PaddlePaddle Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_paddle.py</span></code>)</p></li>
-<li><p><strong>01:01.729</strong>: <a class="reference internal" href="from_tensorflow.html#sphx-glr-how-to-compile-models-from-tensorflow-py"><span class="std std-ref">Compile Tensorflow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tensorflow.py</span></code>)</p></li>
-<li><p><strong>00:57.601</strong>: <a class="reference internal" href="from_darknet.html#sphx-glr-how-to-compile-models-from-darknet-py"><span class="std std-ref">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_darknet.py</span></code>)</p></li>
-<li><p><strong>00:25.458</strong>: <a class="reference internal" href="from_tflite.html#sphx-glr-how-to-compile-models-from-tflite-py"><span class="std std-ref">Compile TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tflite.py</span></code>)</p></li>
-<li><p><strong>00:22.648</strong>: <a class="reference internal" href="from_coreml.html#sphx-glr-how-to-compile-models-from-coreml-py"><span class="std std-ref">Compile CoreML Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_coreml.py</span></code>)</p></li>
-<li><p><strong>00:21.624</strong>: <a class="reference internal" href="from_mxnet.html#sphx-glr-how-to-compile-models-from-mxnet-py"><span class="std std-ref">Compile MXNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_mxnet.py</span></code>)</p></li>
-<li><p><strong>00:19.162</strong>: <a class="reference internal" href="from_pytorch.html#sphx-glr-how-to-compile-models-from-pytorch-py"><span class="std std-ref">Compile PyTorch Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_pytorch.py</span></code>)</p></li>
-<li><p><strong>00:13.923</strong>: <a class="reference internal" href="from_keras.html#sphx-glr-how-to-compile-models-from-keras-py"><span class="std std-ref">Compile Keras Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_keras.py</span></code>)</p></li>
-<li><p><strong>00:02.536</strong>: <a class="reference internal" href="from_onnx.html#sphx-glr-how-to-compile-models-from-onnx-py"><span class="std std-ref">Compile ONNX Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_onnx.py</span></code>)</p></li>
+<li><p><strong>01:11.807</strong>: <a class="reference internal" href="from_paddle.html#sphx-glr-how-to-compile-models-from-paddle-py"><span class="std std-ref">Compile PaddlePaddle Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_paddle.py</span></code>)</p></li>
+<li><p><strong>01:03.238</strong>: <a class="reference internal" href="from_tensorflow.html#sphx-glr-how-to-compile-models-from-tensorflow-py"><span class="std std-ref">Compile Tensorflow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tensorflow.py</span></code>)</p></li>
+<li><p><strong>00:55.430</strong>: <a class="reference internal" href="from_darknet.html#sphx-glr-how-to-compile-models-from-darknet-py"><span class="std std-ref">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_darknet.py</span></code>)</p></li>
+<li><p><strong>00:35.096</strong>: <a class="reference internal" href="from_oneflow.html#sphx-glr-how-to-compile-models-from-oneflow-py"><span class="std std-ref">Compile OneFlow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_oneflow.py</span></code>)</p></li>
+<li><p><strong>00:25.403</strong>: <a class="reference internal" href="from_tflite.html#sphx-glr-how-to-compile-models-from-tflite-py"><span class="std std-ref">Compile TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tflite.py</span></code>)</p></li>
+<li><p><strong>00:22.619</strong>: <a class="reference internal" href="from_pytorch.html#sphx-glr-how-to-compile-models-from-pytorch-py"><span class="std std-ref">Compile PyTorch Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_pytorch.py</span></code>)</p></li>
+<li><p><strong>00:21.671</strong>: <a class="reference internal" href="from_mxnet.html#sphx-glr-how-to-compile-models-from-mxnet-py"><span class="std std-ref">Compile MXNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_mxnet.py</span></code>)</p></li>
+<li><p><strong>00:21.390</strong>: <a class="reference internal" href="from_coreml.html#sphx-glr-how-to-compile-models-from-coreml-py"><span class="std std-ref">Compile CoreML Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_coreml.py</span></code>)</p></li>
+<li><p><strong>00:13.578</strong>: <a class="reference internal" href="from_keras.html#sphx-glr-how-to-compile-models-from-keras-py"><span class="std std-ref">Compile Keras Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_keras.py</span></code>)</p></li>
+<li><p><strong>00:02.679</strong>: <a class="reference internal" href="from_onnx.html#sphx-glr-how-to-compile-models-from-onnx-py"><span class="std std-ref">Compile ONNX Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_onnx.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/deploy/index.html b/docs/how_to/deploy/index.html
index 6f3563c1c..482c71316 100644
--- a/docs/how_to/deploy/index.html
+++ b/docs/how_to/deploy/index.html
@@ -46,7 +46,7 @@
<link rel="index" title="Index" href="../../genindex.html" />
<link rel="search" title="Search" href="../../search.html" />
<link rel="next" title="Deploy TVM Module using C++ API" href="cpp_deploy.html" />
- <link rel="prev" title="Compile PaddlePaddle Models" href="../compile_models/from_paddle.html" />
+ <link rel="prev" title="Compile OneFlow Models" href="../compile_models/from_oneflow.html" />
</head>
<body class="wy-body-for-nav">
@@ -548,7 +548,7 @@ describe how to prepare and deploy models to many of the supported backends.</p>
<a href="cpp_deploy.html" class="btn btn-neutral float-right" title="Deploy TVM Module using C++ API" accesskey="n" rel="next">Next <span class="fa fa-arrow-circle-right"></span></a>
- <a href="../compile_models/from_paddle.html" class="btn btn-neutral float-left" title="Compile PaddlePaddle Models" accesskey="p" rel="prev"><span class="fa fa-arrow-circle-left"></span> Previous</a>
+ <a href="../compile_models/from_oneflow.html" class="btn btn-neutral float-left" title="Compile OneFlow Models" accesskey="p" rel="prev"><span class="fa fa-arrow-circle-left"></span> Previous</a>
</div>
diff --git a/docs/how_to/deploy_models/deploy_model_on_android.html b/docs/how_to/deploy_models/deploy_model_on_android.html
index 24c9d4714..0c368db63 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -622,7 +622,7 @@ to the remote android device.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 16.0768 16.0997 16.1869 15.9239 0.0693
+ 15.5388 15.5206 15.6660 15.4596 0.0746
</pre></div>
</div>
</div>
diff --git a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
index acd095b3a..6fe0ceb71 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -409,14 +409,102 @@ be unstable.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
0%| | 0.00/170M [00:00<?, ?B/s]
- 9%|9 | 15.5M/170M [00:00<00:00, 163MB/s]
- 22%|##2 | 37.7M/170M [00:00<00:00, 204MB/s]
- 36%|###6 | 61.6M/170M [00:00<00:00, 225MB/s]
- 50%|####9 | 84.7M/170M [00:00<00:00, 232MB/s]
- 64%|######3 | 108M/170M [00:00<00:00, 237MB/s]
- 77%|#######7 | 131M/170M [00:00<00:00, 239MB/s]
- 92%|#########2| 157M/170M [00:00<00:00, 249MB/s]
-100%|##########| 170M/170M [00:00<00:00, 237MB/s]
+ 1%| | 0.99M/170M [00:00<00:17, 9.93MB/s]
+ 1%|1 | 2.41M/170M [00:00<00:13, 12.7MB/s]
+ 2%|2 | 3.63M/170M [00:00<00:16, 10.8MB/s]
+ 3%|3 | 5.10M/170M [00:00<00:14, 12.2MB/s]
+ 4%|3 | 6.29M/170M [00:00<00:15, 11.4MB/s]
+ 5%|4 | 8.35M/170M [00:00<00:11, 14.4MB/s]
+ 6%|6 | 10.6M/170M [00:00<00:09, 17.2MB/s]
+ 7%|7 | 12.3M/170M [00:00<00:10, 16.0MB/s]
+ 8%|8 | 13.9M/170M [00:01<00:11, 14.6MB/s]
+ 9%|9 | 15.3M/170M [00:01<00:11, 14.0MB/s]
+ 10%|9 | 16.7M/170M [00:01<00:13, 11.9MB/s]
+ 11%|# | 18.4M/170M [00:01<00:12, 13.2MB/s]
+ 13%|#3 | 22.3M/170M [00:01<00:07, 20.1MB/s]
+ 14%|#4 | 24.4M/170M [00:01<00:07, 20.5MB/s]
+ 16%|#5 | 26.4M/170M [00:01<00:08, 17.7MB/s]
+ 17%|#6 | 28.3M/170M [00:01<00:08, 17.2MB/s]
+ 18%|#7 | 30.1M/170M [00:02<00:09, 16.3MB/s]
+ 19%|#8 | 32.1M/170M [00:02<00:08, 17.4MB/s]
+ 20%|#9 | 33.8M/170M [00:02<00:09, 14.7MB/s]
+ 21%|## | 35.3M/170M [00:02<00:11, 12.8MB/s]
+ 22%|##1 | 36.6M/170M [00:02<00:11, 12.5MB/s]
+ 22%|##2 | 37.9M/170M [00:02<00:11, 11.7MB/s]
+ 23%|##3 | 39.5M/170M [00:02<00:10, 12.9MB/s]
+ 24%|##4 | 41.3M/170M [00:02<00:09, 14.3MB/s]
+ 25%|##5 | 42.8M/170M [00:03<00:10, 12.7MB/s]
+ 26%|##6 | 44.5M/170M [00:03<00:09, 14.0MB/s]
+ 27%|##7 | 45.9M/170M [00:03<00:09, 13.6MB/s]
+ 28%|##8 | 47.7M/170M [00:03<00:08, 14.9MB/s]
+ 29%|##9 | 49.5M/170M [00:03<00:07, 15.8MB/s]
+ 30%|### | 51.1M/170M [00:03<00:08, 15.0MB/s]
+ 31%|### | 52.5M/170M [00:03<00:08, 15.1MB/s]
+ 32%|###1 | 54.1M/170M [00:03<00:07, 15.3MB/s]
+ 33%|###2 | 55.6M/170M [00:03<00:07, 15.6MB/s]
+ 34%|###3 | 57.2M/170M [00:04<00:07, 15.1MB/s]
+ 35%|###4 | 59.1M/170M [00:04<00:07, 16.3MB/s]
+ 36%|###6 | 61.2M/170M [00:04<00:06, 17.6MB/s]
+ 37%|###7 | 62.9M/170M [00:04<00:06, 16.0MB/s]
+ 38%|###7 | 64.5M/170M [00:04<00:07, 14.5MB/s]
+ 39%|###9 | 67.0M/170M [00:04<00:06, 17.5MB/s]
+ 41%|#### | 69.0M/170M [00:04<00:05, 18.2MB/s]
+ 42%|####2 | 71.6M/170M [00:04<00:04, 20.7MB/s]
+ 43%|####3 | 73.7M/170M [00:05<00:05, 19.7MB/s]
+ 45%|####4 | 75.6M/170M [00:05<00:05, 16.8MB/s]
+ 46%|####6 | 78.2M/170M [00:05<00:04, 19.5MB/s]
+ 47%|####7 | 80.2M/170M [00:05<00:04, 18.9MB/s]
+ 48%|####8 | 82.1M/170M [00:05<00:05, 18.1MB/s]
+ 49%|####9 | 83.9M/170M [00:05<00:05, 18.0MB/s]
+ 50%|##### | 85.7M/170M [00:05<00:05, 15.7MB/s]
+ 51%|#####1 | 87.2M/170M [00:05<00:05, 15.1MB/s]
+ 52%|#####2 | 88.7M/170M [00:06<00:06, 13.6MB/s]
+ 53%|#####3 | 90.1M/170M [00:06<00:06, 13.6MB/s]
+ 54%|#####3 | 91.7M/170M [00:06<00:05, 14.3MB/s]
+ 55%|#####4 | 93.1M/170M [00:06<00:05, 13.6MB/s]
+ 56%|#####5 | 94.8M/170M [00:06<00:05, 14.6MB/s]
+ 57%|#####6 | 96.5M/170M [00:06<00:04, 15.4MB/s]
+ 58%|#####7 | 98.3M/170M [00:06<00:04, 16.2MB/s]
+ 59%|#####8 | 99.9M/170M [00:06<00:05, 14.2MB/s]
+ 60%|#####9 | 102M/170M [00:06<00:04, 15.7MB/s]
+ 61%|###### | 103M/170M [00:07<00:05, 12.7MB/s]
+ 62%|######2 | 105M/170M [00:07<00:04, 14.4MB/s]
+ 63%|######2 | 107M/170M [00:07<00:05, 12.5MB/s]
+ 64%|######4 | 109M/170M [00:07<00:04, 14.3MB/s]
+ 65%|######4 | 110M/170M [00:07<00:04, 14.4MB/s]
+ 66%|######6 | 112M/170M [00:07<00:03, 15.8MB/s]
+ 67%|######7 | 114M/170M [00:07<00:03, 16.0MB/s]
+ 68%|######8 | 116M/170M [00:07<00:03, 16.5MB/s]
+ 69%|######9 | 117M/170M [00:08<00:03, 15.3MB/s]
+ 70%|######9 | 119M/170M [00:08<00:03, 15.6MB/s]
+ 71%|####### | 120M/170M [00:08<00:03, 15.6MB/s]
+ 72%|#######1 | 122M/170M [00:08<00:03, 15.9MB/s]
+ 73%|#######2 | 124M/170M [00:08<00:03, 16.1MB/s]
+ 74%|#######4 | 126M/170M [00:08<00:02, 17.2MB/s]
+ 75%|#######5 | 127M/170M [00:08<00:02, 15.0MB/s]
+ 76%|#######5 | 129M/170M [00:08<00:02, 15.2MB/s]
+ 77%|#######6 | 130M/170M [00:09<00:03, 11.3MB/s]
+ 78%|#######7 | 132M/170M [00:09<00:03, 12.5MB/s]
+ 79%|#######8 | 134M/170M [00:09<00:02, 14.5MB/s]
+ 80%|######## | 136M/170M [00:09<00:02, 16.0MB/s]
+ 81%|########1 | 138M/170M [00:09<00:02, 15.5MB/s]
+ 82%|########2 | 140M/170M [00:09<00:01, 16.9MB/s]
+ 83%|########3 | 142M/170M [00:09<00:01, 15.6MB/s]
+ 84%|########4 | 143M/170M [00:09<00:01, 15.6MB/s]
+ 85%|########5 | 145M/170M [00:10<00:02, 13.0MB/s]
+ 87%|########6 | 147M/170M [00:10<00:01, 15.8MB/s]
+ 88%|########7 | 149M/170M [00:10<00:01, 17.1MB/s]
+ 89%|########8 | 151M/170M [00:10<00:01, 13.0MB/s]
+ 90%|########9 | 152M/170M [00:10<00:01, 14.2MB/s]
+ 91%|######### | 154M/170M [00:10<00:01, 15.2MB/s]
+ 92%|#########1| 156M/170M [00:10<00:01, 13.7MB/s]
+ 93%|#########2| 158M/170M [00:10<00:00, 14.7MB/s]
+ 94%|#########3| 159M/170M [00:11<00:00, 13.5MB/s]
+ 95%|#########5| 162M/170M [00:11<00:00, 16.3MB/s]
+ 97%|#########6| 164M/170M [00:11<00:00, 19.4MB/s]
+ 98%|#########8| 167M/170M [00:11<00:00, 21.5MB/s]
+100%|#########9| 169M/170M [00:11<00:00, 20.4MB/s]
+100%|##########| 170M/170M [00:11<00:00, 15.4MB/s]
/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py:3878: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
for i in range(dim)
/usr/local/lib/python3.7/dist-packages/torchvision/models/detection/anchor_utils.py:127: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
@@ -509,7 +597,7 @@ torchvision rcnn models.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Get 9 valid boxes
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 8.181 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes 9.500 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-object-detection-pytorch-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7795da4b258c8feff986668b95ef57ad/deploy_object_detection_pytorch.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_object_detection_pytorch.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_prequantized.html b/docs/how_to/deploy_models/deploy_prequantized.html
index 56babc734..17b41332a 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -450,9 +450,16 @@ training. Other models require a full post training calibration.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
0%| | 0.00/13.6M [00:00<?, ?B/s]
- 23%|##2 | 3.05M/13.6M [00:00<00:00, 32.0MB/s]
- 45%|####5 | 6.11M/13.6M [00:00<00:00, 27.3MB/s]
-100%|##########| 13.6M/13.6M [00:00<00:00, 54.2MB/s]
+ 7%|7 | 1.01M/13.6M [00:00<00:01, 10.3MB/s]
+ 16%|#6 | 2.19M/13.6M [00:00<00:01, 11.4MB/s]
+ 28%|##8 | 3.85M/13.6M [00:00<00:00, 14.1MB/s]
+ 38%|###8 | 5.20M/13.6M [00:00<00:00, 10.7MB/s]
+ 50%|####9 | 6.75M/13.6M [00:00<00:00, 11.7MB/s]
+ 59%|#####8 | 7.94M/13.6M [00:00<00:00, 11.3MB/s]
+ 71%|####### | 9.59M/13.6M [00:00<00:00, 12.9MB/s]
+ 80%|######## | 10.9M/13.6M [00:00<00:00, 11.8MB/s]
+ 89%|########9 | 12.1M/13.6M [00:01<00:00, 12.0MB/s]
+100%|##########| 13.6M/13.6M [00:01<00:00, 12.2MB/s]
</pre></div>
</div>
</div>
@@ -541,7 +548,7 @@ output values are identical out of 1000 outputs from mobilenet v2.</p>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 88.7699 88.7508 92.6485 87.9840 0.6470
+ 90.0302 89.9670 91.2822 89.8079 0.2209
</pre></div>
</div>
<div class="admonition note">
@@ -580,7 +587,7 @@ This includes support for the VNNI 8 bit dot product instruction (CascadeLake or
<div class="section" id="deploy-a-quantized-tflite-model">
<h2>Deploy a quantized TFLite Model<a class="headerlink" href="#deploy-a-quantized-tflite-model" title="Permalink to this headline">¶</a></h2>
<p>TODO</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 5.984 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 4.515 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-prequantized-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/fb8217c13f4351224c6cf3aacf1a87fc/deploy_prequantized.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_prequantized.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_prequantized_tflite.html b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
index 17d4eb1a8..e542ab273 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -540,7 +540,7 @@ TFLite Top-5 labels: [387 102 386 341 349]
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 121.0073 120.9071 122.0060 120.2682 0.4645
+ 118.0240 117.8995 122.2729 116.1341 0.9010
</pre></div>
</div>
<div class="admonition note">
@@ -568,7 +568,7 @@ network for ARM CPU</span></a>.</p></li>
</ul>
</div></blockquote>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 52.710 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 52.119 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-prequantized-tflite-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/56691c7a27d45da61d112276334640d3/deploy_prequantized_tflite.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_prequantized_tflite.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_quantized.html b/docs/how_to/deploy_models/deploy_quantized.html
index 1c0072321..099ec8502 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -480,7 +480,7 @@ for calibration. But the accuracy might be impacted.</p>
DeprecationWarning,
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 9.701 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 36.736 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-quantized-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/7810ecf51bfc05f7d5e8a400ac3e815d/deploy_quantized.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_quantized.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
index f19da32f1..63b92ef3f 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -415,23 +415,23 @@ to your device.</p>
Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
0%| | 0/132723 [00:00<?, ?KB/s]
- 1%|1 | 1479/132723 [00:00<00:09, 14378.80KB/s]
- 4%|3 | 4950/132723 [00:00<00:04, 25952.63KB/s]
- 8%|8 | 10996/132723 [00:00<00:02, 41525.98KB/s]
- 14%|#4 | 18751/132723 [00:00<00:02, 55645.45KB/s]
- 21%|## | 27503/132723 [00:00<00:01, 67083.48KB/s]
- 27%|##7 | 36367/132723 [00:00<00:01, 74388.69KB/s]
- 34%|###4 | 45201/132723 [00:00<00:01, 78937.14KB/s]
- 41%|#### | 54041/132723 [00:00<00:00, 81938.85KB/s]
- 47%|####7 | 63017/132723 [00:00<00:00, 84376.82KB/s]
- 54%|#####4 | 71905/132723 [00:01<00:00, 85764.19KB/s]
- 61%|###### | 80832/132723 [00:01<00:00, 86833.18KB/s]
- 68%|######7 | 89654/132723 [00:01<00:00, 87252.39KB/s]
- 74%|#######4 | 98382/132723 [00:01<00:00, 82998.12KB/s]
- 80%|######## | 106743/132723 [00:01<00:00, 83173.92KB/s]
- 87%|########7 | 115536/132723 [00:01<00:00, 84570.54KB/s]
- 94%|#########3| 124404/132723 [00:01<00:00, 85782.74KB/s]
-100%|##########| 132723/132723 [00:01<00:00, 77340.97KB/s]
+ 5%|4 | 6368/132723 [00:00<00:01, 63672.32KB/s]
+ 11%|#1 | 14897/132723 [00:00<00:01, 76385.98KB/s]
+ 18%|#7 | 23479/132723 [00:00<00:01, 80689.22KB/s]
+ 24%|##4 | 32087/132723 [00:00<00:01, 82814.23KB/s]
+ 31%|### | 40770/132723 [00:00<00:01, 84259.53KB/s]
+ 37%|###7 | 49384/132723 [00:00<00:00, 84894.82KB/s]
+ 44%|####3 | 57945/132723 [00:00<00:00, 85125.32KB/s]
+ 50%|##### | 66458/132723 [00:00<00:00, 71183.84KB/s]
+ 56%|#####5 | 73939/132723 [00:01<00:00, 60487.81KB/s]
+ 61%|###### | 80456/132723 [00:01<00:00, 60996.03KB/s]
+ 67%|######6 | 88391/132723 [00:01<00:00, 65757.99KB/s]
+ 73%|#######3 | 96963/132723 [00:01<00:00, 71150.09KB/s]
+ 80%|#######9 | 105644/132723 [00:01<00:00, 75512.47KB/s]
+ 86%|########6 | 114291/132723 [00:01<00:00, 78627.39KB/s]
+ 93%|#########2| 122937/132723 [00:01<00:00, 80888.66KB/s]
+ 99%|#########9| 131597/132723 [00:01<00:00, 82556.80KB/s]
+100%|##########| 132723/132723 [00:01<00:00, 75800.40KB/s]
</pre></div>
</div>
<p>Create TVM runtime and do inference
@@ -471,7 +471,7 @@ Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from h
</pre></div>
</div>
<img alt="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" class="sphx-glr-single-img" src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" />
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 24.945 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 19.867 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-ssd-gluoncv-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/cccb17d28e5e8b2e94ea8cd5ec59f6ed/deploy_ssd_gluoncv.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_ssd_gluoncv.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/sg_execution_times.html b/docs/how_to/deploy_models/sg_execution_times.html
index c52d3eae2..d31484618 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -300,15 +300,15 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-deploy-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>10:31.937</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>10:52.604</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
<ul class="simple">
-<li><p><strong>03:08.181</strong>: <a class="reference internal" href="deploy_object_detection_pytorch.html#sphx-glr-how-to-deploy-models-deploy-object-detection-pytorch-py"><span class="std std-ref">Compile PyTorch Object Detection Models</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_object_detection_pytorch.py</span></code>)</p></li>
-<li><p><strong>02:24.945</strong>: <a class="reference internal" href="deploy_ssd_gluoncv.html#sphx-glr-how-to-deploy-models-deploy-ssd-gluoncv-py"><span class="std std-ref">Deploy Single Shot Multibox Detector(SSD) model</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_ssd_gluoncv.py</span></code>)</p></li>
-<li><p><strong>01:52.710</strong>: <a class="reference internal" href="deploy_prequantized_tflite.html#sphx-glr-how-to-deploy-models-deploy-prequantized-tflite-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM - Part 3 (TFLite)</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized_tflite.py</span></code>)</p></li>
-<li><p><strong>01:09.701</strong>: <a class="reference internal" href="deploy_quantized.html#sphx-glr-how-to-deploy-models-deploy-quantized-py"><span class="std std-ref">Deploy a Quantized Model on Cuda</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_quantized.py</span></code>)</p></li>
-<li><p><strong>01:05.984</strong>: <a class="reference internal" href="deploy_prequantized.html#sphx-glr-how-to-deploy-models-deploy-prequantized-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized.py</span></code>)</p></li>
-<li><p><strong>00:28.405</strong>: <a class="reference internal" href="deploy_model_on_android.html#sphx-glr-how-to-deploy-models-deploy-model-on-android-py"><span class="std std-ref">Deploy the Pretrained Model on Android</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_android.py</span></code>)</p></li>
-<li><p><strong>00:21.821</strong>: <a class="reference internal" href="deploy_model_on_rasp.html#sphx-glr-how-to-deploy-models-deploy-model-on-rasp-py"><span class="std std-ref">Deploy the Pretrained Model on Raspberry Pi</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_rasp.py</span></code>)</p></li>
+<li><p><strong>03:09.500</strong>: <a class="reference internal" href="deploy_object_detection_pytorch.html#sphx-glr-how-to-deploy-models-deploy-object-detection-pytorch-py"><span class="std std-ref">Compile PyTorch Object Detection Models</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_object_detection_pytorch.py</span></code>)</p></li>
+<li><p><strong>02:19.867</strong>: <a class="reference internal" href="deploy_ssd_gluoncv.html#sphx-glr-how-to-deploy-models-deploy-ssd-gluoncv-py"><span class="std std-ref">Deploy Single Shot Multibox Detector(SSD) model</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_ssd_gluoncv.py</span></code>)</p></li>
+<li><p><strong>01:52.119</strong>: <a class="reference internal" href="deploy_prequantized_tflite.html#sphx-glr-how-to-deploy-models-deploy-prequantized-tflite-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM - Part 3 (TFLite)</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized_tflite.py</span></code>)</p></li>
+<li><p><strong>01:36.736</strong>: <a class="reference internal" href="deploy_quantized.html#sphx-glr-how-to-deploy-models-deploy-quantized-py"><span class="std std-ref">Deploy a Quantized Model on Cuda</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_quantized.py</span></code>)</p></li>
+<li><p><strong>01:04.515</strong>: <a class="reference internal" href="deploy_prequantized.html#sphx-glr-how-to-deploy-models-deploy-prequantized-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized.py</span></code>)</p></li>
+<li><p><strong>00:28.540</strong>: <a class="reference internal" href="deploy_model_on_android.html#sphx-glr-how-to-deploy-models-deploy-model-on-android-py"><span class="std std-ref">Deploy the Pretrained Model on Android</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_android.py</span></code>)</p></li>
+<li><p><strong>00:21.137</strong>: <a class="reference internal" href="deploy_model_on_rasp.html#sphx-glr-how-to-deploy-models-deploy-model-on-rasp-py"><span class="std std-ref">Deploy the Pretrained Model on Raspberry Pi</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_rasp.py</span></code>)</p></li>
<li><p><strong>00:00.190</strong>: <a class="reference internal" href="deploy_sparse.html#sphx-glr-how-to-deploy-models-deploy-sparse-py"><span class="std std-ref">Deploy a Hugging Face Pruned Model on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_sparse.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/extend_tvm/bring_your_own_datatypes.html b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
index c438264b8..9b2fe06d6 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -588,7 +588,7 @@ In this alpha state of the Bring Your Own Datatypes framework, we have not imple
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zipd92d2127-e27d-4c49-b468-f6776fd76cd2 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zipebb22811-12a9-49a8-a8d7-800ae10725cb from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
</pre></div>
</div>
<p>It’s easy to execute MobileNet with native TVM:</p>
@@ -650,7 +650,7 @@ In this alpha state of the Bring Your Own Datatypes framework, we have not imple
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Check failed: (lower) is false: FloatImm lowering function for target llvm type 150 not found
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Check failed: (lower) is false: Intrinsic lowering function for target llvm, intrinsic name tir.sqrt, type 150 not found
</pre></div>
</div>
<p>When we attempt to run the model, we get a familiar error telling us that more functions need to be registerd for myfloat.</p>
diff --git a/docs/how_to/extend_tvm/sg_execution_times.html b/docs/how_to/extend_tvm/sg_execution_times.html
index 0c931e380..ec54ba485 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -300,12 +300,12 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-extend-tvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:38.042</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:39.769</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:34.591</strong>: <a class="reference internal" href="bring_your_own_datatypes.html#sphx-glr-how-to-extend-tvm-bring-your-own-datatypes-py"><span class="std std-ref">Bring Your Own Datatypes to TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">bring_your_own_datatypes.py</span></code>)</p></li>
-<li><p><strong>00:02.217</strong>: <a class="reference internal" href="use_pass_instrument.html#sphx-glr-how-to-extend-tvm-use-pass-instrument-py"><span class="std std-ref">How to Use TVM Pass Instrument</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_instrument.py</span></code>)</p></li>
-<li><p><strong>00:01.032</strong>: <a class="reference internal" href="use_pass_infra.html#sphx-glr-how-to-extend-tvm-use-pass-infra-py"><span class="std std-ref">How to Use TVM Pass Infra</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_infra.py</span></code>)</p></li>
-<li><p><strong>00:00.202</strong>: <a class="reference internal" href="low_level_custom_pass.html#sphx-glr-how-to-extend-tvm-low-level-custom-pass-py"><span class="std std-ref">Writing a Customized Pass</span></a> (<code class="docutils literal notranslate"><span class="pre">low_level_custom_pass.py</span></code>)</p></li>
+<li><p><strong>00:36.149</strong>: <a class="reference internal" href="bring_your_own_datatypes.html#sphx-glr-how-to-extend-tvm-bring-your-own-datatypes-py"><span class="std std-ref">Bring Your Own Datatypes to TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">bring_your_own_datatypes.py</span></code>)</p></li>
+<li><p><strong>00:02.336</strong>: <a class="reference internal" href="use_pass_instrument.html#sphx-glr-how-to-extend-tvm-use-pass-instrument-py"><span class="std std-ref">How to Use TVM Pass Instrument</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_instrument.py</span></code>)</p></li>
+<li><p><strong>00:01.084</strong>: <a class="reference internal" href="use_pass_infra.html#sphx-glr-how-to-extend-tvm-use-pass-infra-py"><span class="std std-ref">How to Use TVM Pass Infra</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_infra.py</span></code>)</p></li>
+<li><p><strong>00:00.200</strong>: <a class="reference internal" href="low_level_custom_pass.html#sphx-glr-how-to-extend-tvm-low-level-custom-pass-py"><span class="std std-ref">Writing a Customized Pass</span></a> (<code class="docutils literal notranslate"><span class="pre">low_level_custom_pass.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index e43844609..0d839be3a 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -486,10 +486,10 @@ profile the execution time of each passes.</p>
</div>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 5919us [5919us] (45.59%; 45.59%)
-FoldScaleAxis: 7065us [2us] (54.41%; 54.41%)
- FoldConstant: 7063us [1481us] (54.40%; 99.97%)
- InferType: 5581us [5581us] (42.99%; 79.03%)
+InferType: 6301us [6301us] (45.44%; 45.44%)
+FoldScaleAxis: 7566us [2us] (54.56%; 54.56%)
+ FoldConstant: 7564us [1556us] (54.55%; 99.97%)
+ InferType: 6008us [6008us] (43.33%; 79.43%)
</pre></div>
</div>
</div>
@@ -512,10 +512,10 @@ Refer to following sections and <a class="reference internal" href="../../refere
</div>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 5642us [5642us] (44.48%; 44.48%)
-FoldScaleAxis: 7044us [2us] (55.52%; 55.52%)
- FoldConstant: 7042us [1448us] (55.51%; 99.97%)
- InferType: 5594us [5594us] (44.09%; 79.44%)
+InferType: 6063us [6063us] (44.71%; 44.71%)
+FoldScaleAxis: 7497us [2us] (55.29%; 55.29%)
+ FoldConstant: 7495us [1531us] (55.27%; 99.97%)
+ InferType: 5964us [5964us] (43.98%; 79.57%)
</pre></div>
</div>
<p>Register empty list to clear existing instruments.</p>
diff --git a/docs/how_to/optimize_operators/opt_conv_cuda.html b/docs/how_to/optimize_operators/opt_conv_cuda.html
index 32114cdfa..b9244590d 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -534,7 +534,7 @@ latency of convolution.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 54.266551 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 47.082751 ms
</pre></div>
</div>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-optimize-operators-opt-conv-cuda-py">
diff --git a/docs/how_to/optimize_operators/opt_conv_tensorcore.html b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
index 63395b946..5a5f9547b 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -878,7 +878,7 @@ be able to run on our build server</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 7.173443 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 6.855349 ms
</pre></div>
</div>
</div>
diff --git a/docs/how_to/optimize_operators/opt_gemm.html b/docs/how_to/optimize_operators/opt_gemm.html
index 97de33617..0910e26c2 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -431,8 +431,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.019053
-Baseline: 3.195089
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018223
+Baseline: 3.524503
</pre></div>
</div>
<p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -494,7 +494,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.298429
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.297080
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -563,7 +563,7 @@ vastly.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.335815
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.328934
</pre></div>
</div>
<p>Here is the generated IR after vectorization.</p>
@@ -626,7 +626,7 @@ the access pattern for A matrix is more cache friendly.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.116504
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.114862
</pre></div>
</div>
<p>Here is the generated IR after loop permutation.</p>
@@ -711,7 +711,7 @@ flattening.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.109173
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.111236
</pre></div>
</div>
<p>Here is the generated IR after array packing.</p>
@@ -799,7 +799,7 @@ write to C when all the block results are ready.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.109061
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111643
</pre></div>
</div>
<p>Here is the generated IR after blocking.</p>
@@ -891,7 +891,7 @@ write to C when all the block results are ready.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.141408
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.144929
</pre></div>
</div>
<p>Here is the generated IR after parallelization.</p>
diff --git a/docs/how_to/optimize_operators/sg_execution_times.html b/docs/how_to/optimize_operators/sg_execution_times.html
index e4c64c287..bbfbc43ce 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -300,11 +300,11 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-optimize-operators-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:34.103</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:35.179</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:31.489</strong>: <a class="reference internal" href="opt_gemm.html#sphx-glr-how-to-optimize-operators-opt-gemm-py"><span class="std std-ref">How to optimize GEMM on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_gemm.py</span></code>)</p></li>
-<li><p><strong>00:01.395</strong>: <a class="reference internal" href="opt_conv_tensorcore.html#sphx-glr-how-to-optimize-operators-opt-conv-tensorcore-py"><span class="std std-ref">How to optimize convolution using TensorCores</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_conv_tensorcore.py</span></code>)</p></li>
-<li><p><strong>00:01.219</strong>: <a class="reference internal" href="opt_conv_cuda.html#sphx-glr-how-to-optimize-operators-opt-conv-cuda-py"><span class="std std-ref">How to optimize convolution on GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_conv_cuda.py</span></code>)</p></li>
+<li><p><strong>00:32.555</strong>: <a class="reference internal" href="opt_gemm.html#sphx-glr-how-to-optimize-operators-opt-gemm-py"><span class="std std-ref">How to optimize GEMM on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_gemm.py</span></code>)</p></li>
+<li><p><strong>00:01.396</strong>: <a class="reference internal" href="opt_conv_tensorcore.html#sphx-glr-how-to-optimize-operators-opt-conv-tensorcore-py"><span class="std std-ref">How to optimize convolution using TensorCores</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_conv_tensorcore.py</span></code>)</p></li>
+<li><p><strong>00:01.228</strong>: <a class="reference internal" href="opt_conv_cuda.html#sphx-glr-how-to-optimize-operators-opt-conv-cuda-py"><span class="std std-ref">How to optimize convolution on GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_conv_cuda.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
index 04226451a..6e7e3eaac 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -300,14 +300,14 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autoscheduler-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>05:14.315</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>05:02.667</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
<ul class="simple">
-<li><p><strong>02:39.840</strong>: <a class="reference internal" href="tune_conv2d_layer_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py"><span class="std std-ref">Auto-scheduling a Convolution Layer for GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_layer_cuda.py</span></code>)</p></li>
-<li><p><strong>01:18.913</strong>: <a class="reference internal" href="tune_network_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-x86-py"><span class="std std-ref">Auto-scheduling a Neural Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_x86.py</span></code>)</p></li>
-<li><p><strong>00:40.055</strong>: <a class="reference internal" href="tune_network_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-cuda-py"><span class="std std-ref">Auto-scheduling a Neural Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_cuda.py</span></code>)</p></li>
-<li><p><strong>00:18.511</strong>: <a class="reference internal" href="tune_sparse_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-sparse-x86-py"><span class="std std-ref">Auto-scheduling Sparse Matrix Multiplication on CPU with Custom Sketch Rule</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_sparse_x86.py</span></code>)</p></li>
-<li><p><strong>00:08.603</strong>: <a class="reference internal" href="tune_network_mali.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-mali-py"><span class="std std-ref">Auto-scheduling a Neural Network for mali GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_mali.py</span></code>)</p></li>
-<li><p><strong>00:08.392</strong>: <a class="reference internal" href="tune_network_arm.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-arm-py"><span class="std std-ref">Auto-scheduling a Neural Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_arm.py</span></code>)</p></li>
+<li><p><strong>02:27.183</strong>: <a class="reference internal" href="tune_conv2d_layer_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py"><span class="std std-ref">Auto-scheduling a Convolution Layer for GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_layer_cuda.py</span></code>)</p></li>
+<li><p><strong>01:20.913</strong>: <a class="reference internal" href="tune_network_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-x86-py"><span class="std std-ref">Auto-scheduling a Neural Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_x86.py</span></code>)</p></li>
+<li><p><strong>00:40.029</strong>: <a class="reference internal" href="tune_network_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-cuda-py"><span class="std std-ref">Auto-scheduling a Neural Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_cuda.py</span></code>)</p></li>
+<li><p><strong>00:17.815</strong>: <a class="reference internal" href="tune_sparse_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-sparse-x86-py"><span class="std std-ref">Auto-scheduling Sparse Matrix Multiplication on CPU with Custom Sketch Rule</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_sparse_x86.py</span></code>)</p></li>
+<li><p><strong>00:08.462</strong>: <a class="reference internal" href="tune_network_mali.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-mali-py"><span class="std std-ref">Auto-scheduling a Neural Network for mali GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_mali.py</span></code>)</p></li>
+<li><p><strong>00:08.266</strong>: <a class="reference internal" href="tune_network_arm.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-arm-py"><span class="std std-ref">Auto-scheduling a Neural Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_arm.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html b/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
index 7d13df593..537ca496c 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
@@ -450,7 +450,7 @@ file and apply it.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>.T
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>
</pre></div>
</div>
<p>We can lower the schedule to see the IR after auto-scheduling.
@@ -470,12 +470,12 @@ cooperative fetching, unrolling and operator fusion.</p>
compute: Buffer(compute_2: Pointer(float32), float32, [25088], [])}
buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
- attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 128;
- allocate(conv2d_nchw: Pointer(local float32), float32, [28]), storage_scope = local;
- allocate(pad_temp.shared: Pointer(shared float32), float32, [1008]), storage_scope = shared;
- allocate(kernel.shared: Pointer(shared float32), float32, [192]), storage_scope = shared;
- attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7 {
- conv2d_nchw_1: Buffer(conv2d_nchw, float32, [28], [], scope="local", align=64)[0] = 0f32
+ attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 16;
+ allocate(conv2d_nchw: Pointer(local float32), float32, [14]), storage_scope = local;
+ allocate(pad_temp.shared: Pointer(shared float32), float32, [504]), storage_scope = shared;
+ allocate(kernel.shared: Pointer(shared float32), float32, [768]), storage_scope = shared;
+ attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112 {
+ conv2d_nchw_1: Buffer(conv2d_nchw, float32, [14], [], scope="local", align=32)[0] = 0f32
conv2d_nchw_1[7] = 0f32
conv2d_nchw_1[1] = 0f32
conv2d_nchw_1[8] = 0f32
@@ -489,476 +489,77 @@ cooperative fetching, unrolling and operator fusion.</p>
conv2d_nchw_1[12] = 0f32
conv2d_nchw_1[6] = 0f32
conv2d_nchw_1[13] = 0f32
- conv2d_nchw_1[14] = 0f32
- conv2d_nchw_1[21] = 0f32
- conv2d_nchw_1[15] = 0f32
- conv2d_nchw_1[22] = 0f32
- conv2d_nchw_1[16] = 0f32
- conv2d_nchw_1[23] = 0f32
- conv2d_nchw_1[17] = 0f32
- conv2d_nchw_1[24] = 0f32
- conv2d_nchw_1[18] = 0f32
- conv2d_nchw_1[25] = 0f32
- conv2d_nchw_1[19] = 0f32
- conv2d_nchw_1[26] = 0f32
- conv2d_nchw_1[20] = 0f32
- conv2d_nchw_1[27] = 0f32
- for (rc.outer.outer: int32, 0, 32) {
+ for (rc.outer.outer: int32, 0, 64) {
for (rx.outer.outer: int32, 0, 3) {
- let cse_var_1: int32 = (rc.outer.outer*144)
+ let cse_var_2: int32 = (rc.outer.outer*392)
+ let cse_var_1: int32 = (rc.outer.outer*72)
{
- attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1: Buffer(pad_temp.shared, float32, [1008], [], scope="shared")[threadIdx.x_1] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 7)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) - 1)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 14)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 6)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 21)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 13)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 28)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 20)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 35)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 27)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 42)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 34)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 49)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 41)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 56)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 63)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 70)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 48)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 77)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 55)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 84)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 62)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 91)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 69)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 76)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 105)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 83)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 112)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 90)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 119)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 126)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 133)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 97)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 140)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 104)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 147)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 111)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 154)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 118)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 161)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 125)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 168)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 132)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 175)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 139)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 182)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 189)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 146)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 203)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 153)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 210)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 160)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 217)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 167)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 224)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 174)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 231)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 181)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 238)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 188)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 245)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 252)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 259)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 195)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 266)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 202)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 273)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 209)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 280)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 216)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 287)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 223)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 294)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 230)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 301)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 237)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 308)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 315)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 322)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 244)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 329)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 251)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 336)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 258)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 343)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 265)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 350)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 272)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 357)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 279)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 364)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 286)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 371)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 378)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 385)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 293)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 392)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 300)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 399)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 307)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 406)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 314)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 413)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 321)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 420)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 328)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 427)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 335)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 434)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 441)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 448)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 342)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 455)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 349)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 462)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 356)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 469)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 363)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 476)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 370)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 483)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 377)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 490)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 384)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 497)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 504)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 511)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 391)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 518)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 398)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 525)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 405)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 532)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 412)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 539)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 419)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 546)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 426)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 553)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 433)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 560)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 567)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 574)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 440)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 581)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 447)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 588)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 454)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 595)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 461)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 602)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 468)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 609)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 475)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 616)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 482)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 623)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 630)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 637)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 489)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 644)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 496)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 651)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 503)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 658)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 510)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 665)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 517)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 672)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 524)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 679)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 531)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 686)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 693)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 700)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 538)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 707)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 545)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 714)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 552)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 721)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 559)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 728)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 566)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 735)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 573)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 742)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 580)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 749)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 756)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 763)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 587)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 770)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 594)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 777)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 601)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 784)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 608)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 791)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 615)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 798)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 622)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 805)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 629)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 812)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 819)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 826)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 636)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 833)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 643)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 840)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 650)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 847)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 657)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 854)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 664)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 861)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 671)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 868)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 678)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 875)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 882)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 889)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 685)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 896)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 692)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 903)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 699)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 910)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 706)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 917)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 713)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 924)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 720)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 931)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 727)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 938)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 945)] = 0f32
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 952)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 734)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 959)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 741)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 966)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 748)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 973)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 755)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 980)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 762)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 987)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 769)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 994)] = @tir.if_then_else(((1 <= (threadIdx.x_1 + rx.outer.outer)) && ((threadIdx.x_1 + rx.outer.outer) < 8)), data[((((rc.outer.outer*784) + rx.outer.outer) + threadIdx.x_1) + 776)], 0f32, dtype=float32)
- attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- pad_temp.shared_1[(threadIdx.x_1 + 1001)] = 0f32
- attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1: Buffer(kernel.shared, float32, [192], [], scope="shared")[threadIdx.x_2] = kernel[((((blockIdx.x*18432) + cse_var_1) + (threadIdx.x_2*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 7)] = kernel[((((blockIdx.x*18432) + cse_var_1) + ((threadIdx.x_2 + 7)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 14)] = kernel[((((blockIdx.x*18432) + cse_var_1) + ((threadIdx.x_2 + 14)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 21)] = kernel[((((blockIdx.x*18432) + cse_var_1) + ((threadIdx.x_2 + 21)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 28)] = kernel[((((blockIdx.x*18432) + cse_var_1) + ((threadIdx.x_2 + 28)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 35)] = kernel[((((blockIdx.x*18432) + cse_var_1) + ((threadIdx.x_2 + 35)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 42)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 42), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 42), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 49)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 49), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 1), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 56)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 56), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 8), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 63)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 63), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 15), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 70)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 70), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 22), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 77)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 77), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 29), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 84)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 84), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 36), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 91)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 91), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 43), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 98)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 98), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 2), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 105)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 105), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 9), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 112), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 16), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 119)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 119), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 23), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 126)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 126), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 30), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 133)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 133), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 37), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 140)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 140), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 44), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 147)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 147), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 3), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 154)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 154), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 10), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 161)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 161), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 17), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 168)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 168), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 24), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 175)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 175), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 31), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- kernel.shared_1[(threadIdx.x_2 + 182)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 182), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 38), 48)*3)) + rx.outer.outer)]
- attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 7;
- if @tir.likely((threadIdx.x_2 < 3), dtype=bool) {
- kernel.shared_1[(threadIdx.x_2 + 189)] = kernel[(((((blockIdx.x*18432) + (floordiv((threadIdx.x_2 + 189), 48)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 45), 48)*3)) + rx.outer.outer)]
+ attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ pad_temp.shared_1: Buffer(pad_temp.shared, float32, [504], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else(((((7 <= floormod(threadIdx.x_1, 63)) && (floormod(threadIdx.x_1, 63) < 56)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[((((cse_var_2 + (floordiv(threadIdx.x_1, 63)*49)) + rx.outer.outer) + floormod(threadIdx.x_1, 63)) - 8)], 0f32, dtype=float32)
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ pad_temp.shared_1[(threadIdx.x_1 + 112)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((floordiv(threadIdx.x_1, 7) + 16), 9)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + rx.outer.outer) + floormod(th [...]
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ pad_temp.shared_1[(threadIdx.x_1 + 224)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((floordiv(threadIdx.x_1, 7) + 32), 9)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + rx.outer.outer) + floormod(th [...]
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ pad_temp.shared_1[(threadIdx.x_1 + 336)] = @tir.if_then_else(((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 3), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) < 8)) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((floordiv(threadIdx.x_1, 7) + 48), 9)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + rx.outer.outer) + floormod(th [...]
+ attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ if @tir.likely((threadIdx.x_1 < 56), dtype=bool) {
+ pad_temp.shared_1[(threadIdx.x_1 + 448)] = @tir.if_then_else((((floormod((floordiv(threadIdx.x_1, 7) + 1), 9) < 8) && (1 <= (rx.outer.outer + floormod(threadIdx.x_1, 7)))) && ((rx.outer.outer + floormod(threadIdx.x_1, 7)) < 8)), data[(((((cse_var_2 + (floordiv((floordiv(threadIdx.x_1, 7) + 64), 9)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 1), 9)*7)) + rx.outer.outer) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
}
- for (rc.outer.inner: int32, 0, 16) {
- let cse_var_13: int32 = (rc.outer.inner*3)
- let cse_var_12: int32 = (cse_var_13 + 1)
- let cse_var_11: int32 = (cse_var_13 + 144)
- let cse_var_10: int32 = (cse_var_13 + 146)
- let cse_var_9: int32 = (cse_var_13 + 2)
- let cse_var_8: int32 = (cse_var_13 + 48)
- let cse_var_7: int32 = (cse_var_13 + 49)
- let cse_var_6: int32 = (cse_var_13 + 50)
- let cse_var_5: int32 = (cse_var_13 + 96)
- let cse_var_4: int32 = (cse_var_13 + 97)
- let cse_var_3: int32 = (cse_var_13 + 98)
- let cse_var_2: int32 = (cse_var_13 + 145)
- {
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((rc.outer.inner*63) + threadIdx.x)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((rc.outer.inner*63) + threadIdx.x)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_13]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_8]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_12]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_7]))
- conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 56)]*kernel.shared_1[cse_var_9]))
- conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 56)]*kernel.shared_1[cse_var_6]))
- conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[((rc.outer.inner*63) + threadIdx.x)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[((rc.outer.inner*63) + threadIdx.x)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[14] = (conv2d_nchw_1[14] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[21] = (conv2d_nchw_1[21] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_10]))
- conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 7)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[15] = (conv2d_nchw_1[15] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[22] = (conv2d_nchw_1[22] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_10]))
- conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 14)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[16] = (conv2d_nchw_1[16] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[23] = (conv2d_nchw_1[23] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_10]))
- conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 21)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[17] = (conv2d_nchw_1[17] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[24] = (conv2d_nchw_1[24] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_10]))
- conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 28)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[18] = (conv2d_nchw_1[18] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[25] = (conv2d_nchw_1[25] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_10]))
- conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 35)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[19] = (conv2d_nchw_1[19] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[26] = (conv2d_nchw_1[26] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_10]))
- conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_5]))
- conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 42)]*kernel.shared_1[cse_var_11]))
- conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_4]))
- conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 49)]*kernel.shared_1[cse_var_2]))
- conv2d_nchw_1[20] = (conv2d_nchw_1[20] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 56)]*kernel.shared_1[cse_var_3]))
- conv2d_nchw_1[27] = (conv2d_nchw_1[27] + (pad_temp.shared_1[(((rc.outer.inner*63) + threadIdx.x) + 56)]*kernel.shared_1[cse_var_10]))
+ attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ kernel.shared_1: Buffer(kernel.shared, float32, [768], [], scope="shared")[threadIdx.x_2] = kernel[(((((blockIdx.x*147456) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_1) + (floormod(threadIdx.x_2, 24)*3)) + rx.outer.outer)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ kernel.shared_1[(threadIdx.x_2 + 112)] = kernel[(((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 8) + 14), 3)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 16), 24)*3)) + rx.outer.outer)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ kernel.shared_1[(threadIdx.x_2 + 224)] = kernel[(((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 8) + 28), 3)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 8), 24)*3)) + rx.outer.outer)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ kernel.shared_1[(threadIdx.x_2 + 336)] = kernel[((((((blockIdx.x*147456) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_1) + (floormod(threadIdx.x_2, 24)*3)) + rx.outer.outer) + 64512)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ kernel.shared_1[(threadIdx.x_2 + 448)] = kernel[(((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 8) + 56), 3)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 16), 24)*3)) + rx.outer.outer)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ kernel.shared_1[(threadIdx.x_2 + 560)] = kernel[(((((blockIdx.x*147456) + (floordiv((floordiv(threadIdx.x_2, 8) + 70), 3)*4608)) + cse_var_1) + (floormod((threadIdx.x_2 + 8), 24)*3)) + rx.outer.outer)]
+ attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 112;
+ if @tir.likely((threadIdx.x_2 < 96), dtype=bool) {
+ kernel.shared_1[(threadIdx.x_2 + 672)] = kernel[((((((blockIdx.x*147456) + (floordiv(floordiv(threadIdx.x_2, 8), 3)*4608)) + cse_var_1) + (floormod(threadIdx.x_2, 24)*3)) + rx.outer.outer) + 129024)]
+ }
+ for (rc.outer.inner: int32, 0, 4) {
+ for (ry.outer.inner: int32, 0, 3) {
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7))]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 63)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 7)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 7)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 70)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 70)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 14)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 14)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 77)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 77)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 21)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 21)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 84)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 84)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 28)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 28)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 91)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 35)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 35)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 98)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 98)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 42)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 42)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 24)]))
+ conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 105)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 3)]))
+ conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((((rc.outer.inner*126) + (ry.outer.inner*7)) + floormod(threadIdx.x, 7)) + 105)]*kernel.shared_1[((((floordiv(threadIdx.x, 7)*48) + (rc.outer.inner*6)) + ry.outer.inner) + 27)]))
}
}
}
}
}
- for (i1.inner: int32, 0, 4) {
+ for (i1.inner: int32, 0, 2) {
for (i2.inner: int32, 0, 7) {
- compute[((((blockIdx.x*196) + (i1.inner*49)) + (i2.inner*7)) + threadIdx.x)] = max((conv2d_nchw_1[((i1.inner*7) + i2.inner)] + bias[((blockIdx.x*4) + i1.inner)]), 0f32)
+ compute[(((((blockIdx.x*1568) + (floordiv(threadIdx.x, 7)*98)) + (i1.inner*49)) + (i2.inner*7)) + floormod(threadIdx.x, 7))] = max((conv2d_nchw_1[((i1.inner*7) + i2.inner)] + bias[(((blockIdx.x*32) + (floordiv(threadIdx.x, 7)*2)) + i1.inner)]), 0f32)
}
}
}
@@ -997,7 +598,7 @@ cooperative fetching, unrolling and operator fusion.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.365 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.413 ms
</pre></div>
</div>
</div>
@@ -1028,8 +629,8 @@ conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o
conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=2)
-conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=2)
-conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=1)
+conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
+conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=16)
conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=1)
conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=7)
@@ -1039,18 +640,18 @@ conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, fact
conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
-conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=1)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=16)
-conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=3)
-conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
+conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
+conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
+conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2d_nc [...]
compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
-compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=4)
-compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=1)
+compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
+compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=16)
compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=1)
compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=7)
compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
@@ -1076,14 +677,14 @@ s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=7)
+kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=7)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=112)
s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
-s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 512)
+s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 64)
s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
CUDA source code:
@@ -1101,10 +702,10 @@ CUDA source code:
#define int64_t long long
#define uint64_t unsigned long long
#endif
-extern "C" __global__ void __launch_bounds__(7) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
- float conv2d_nchw[28];
- __shared__ float pad_temp_shared[1008];
- __shared__ float kernel_shared[192];
+extern "C" __global__ void __launch_bounds__(112) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+ float conv2d_nchw[14];
+ __shared__ float pad_temp_shared[504];
+ __shared__ float kernel_shared[768];
conv2d_nchw[0] = 0.000000e+00f;
conv2d_nchw[7] = 0.000000e+00f;
conv2d_nchw[1] = 0.000000e+00f;
@@ -1119,289 +720,63 @@ extern "C" __global__ void __launch_bounds__(7) default_function_kerne
conv2d_nchw[12] = 0.000000e+00f;
conv2d_nchw[6] = 0.000000e+00f;
conv2d_nchw[13] = 0.000000e+00f;
- conv2d_nchw[14] = 0.000000e+00f;
- conv2d_nchw[21] = 0.000000e+00f;
- conv2d_nchw[15] = 0.000000e+00f;
- conv2d_nchw[22] = 0.000000e+00f;
- conv2d_nchw[16] = 0.000000e+00f;
- conv2d_nchw[23] = 0.000000e+00f;
- conv2d_nchw[17] = 0.000000e+00f;
- conv2d_nchw[24] = 0.000000e+00f;
- conv2d_nchw[18] = 0.000000e+00f;
- conv2d_nchw[25] = 0.000000e+00f;
- conv2d_nchw[19] = 0.000000e+00f;
- conv2d_nchw[26] = 0.000000e+00f;
- conv2d_nchw[20] = 0.000000e+00f;
- conv2d_nchw[27] = 0.000000e+00f;
- for (int rc_outer_outer = 0; rc_outer_outer < 32; ++rc_outer_outer) {
+ for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
for (int rx_outer_outer = 0; rx_outer_outer < 3; ++rx_outer_outer) {
__syncthreads();
- pad_temp_shared[((int)threadIdx.x)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 7)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) - 1)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 14)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 6)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 21)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 13)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 28)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 20)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 35)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 27)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 42)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 34)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 49)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 41)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 56)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 63)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 70)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 48)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 77)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 55)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 84)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 62)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 91)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 69)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 98)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 76)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 105)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 83)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 112)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 90)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 119)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 126)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 133)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 97)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 140)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 104)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 147)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 111)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 154)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 118)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 161)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 125)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 168)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 132)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 175)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 139)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 182)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 189)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 196)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 146)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 203)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 153)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 210)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 160)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 217)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 167)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 224)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 174)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 231)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 181)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 238)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 188)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 245)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 252)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 259)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 195)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 266)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 202)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 273)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 209)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 280)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 216)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 287)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 223)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 294)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 230)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 301)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 237)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 308)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 315)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 322)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 244)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 329)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 251)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 336)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 258)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 343)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 265)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 350)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 272)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 357)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 279)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 364)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 286)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 371)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 378)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 385)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 293)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 392)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 300)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 399)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 307)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 406)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 314)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 413)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 321)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 420)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 328)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 427)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 335)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 434)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 441)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 448)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 342)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 455)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 349)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 462)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 356)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 469)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 363)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 476)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 370)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 483)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 377)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 490)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 384)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 497)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 504)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 511)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 391)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 518)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 398)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 525)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 405)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 532)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 412)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 539)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 419)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 546)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 426)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 553)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 433)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 560)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 567)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 574)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 440)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 581)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 447)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 588)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 454)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 595)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 461)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 602)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 468)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 609)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 475)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 616)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 482)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 623)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 630)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 637)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 489)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 644)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 496)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 651)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 503)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 658)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 510)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 665)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 517)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 672)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 524)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 679)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 531)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 686)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 693)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 700)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 538)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 707)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 545)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 714)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 552)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 721)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 559)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 728)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 566)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 735)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 573)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 742)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 580)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 749)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 756)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 763)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 587)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 770)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 594)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 777)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 601)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 784)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 608)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 791)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 615)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 798)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 622)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 805)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 629)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 812)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 819)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 826)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 636)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 833)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 643)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 840)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 650)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 847)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 657)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 854)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 664)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 861)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 671)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 868)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 678)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 875)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 882)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 889)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 685)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 896)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 692)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 903)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 699)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 910)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 706)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 917)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 713)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 924)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 720)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 931)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 727)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 938)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 945)] = 0.000000e+00f;
- pad_temp_shared[(((int)threadIdx.x) + 952)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 734)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 959)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 741)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 966)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 748)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 973)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 755)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 980)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 762)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 987)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 769)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 994)] = (((1 <= (((int)threadIdx.x) + rx_outer_outer)) && ((((int)threadIdx.x) + rx_outer_outer) < 8)) ? data[((((rc_outer_outer * 784) + rx_outer_outer) + ((int)threadIdx.x)) + 776)] : 0.000000e+00f);
- pad_temp_shared[(((int)threadIdx.x) + 1001)] = 0.000000e+00f;
- kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 18432) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 7)] = kernel[(((((((int)blockIdx.x) * 18432) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 21)];
- kernel_shared[(((int)threadIdx.x) + 14)] = kernel[(((((((int)blockIdx.x) * 18432) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 42)];
- kernel_shared[(((int)threadIdx.x) + 21)] = kernel[(((((((int)blockIdx.x) * 18432) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 63)];
- kernel_shared[(((int)threadIdx.x) + 28)] = kernel[(((((((int)blockIdx.x) * 18432) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 84)];
- kernel_shared[(((int)threadIdx.x) + 35)] = kernel[(((((((int)blockIdx.x) * 18432) + (rc_outer_outer * 144)) + (((int)threadIdx.x) * 3)) + rx_outer_outer) + 105)];
- kernel_shared[(((int)threadIdx.x) + 42)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 42) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 42) % 48) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 49)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 49) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 1) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 56)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 56) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 8) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 63)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 63) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 15) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 70)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 70) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 22) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 77)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 77) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 29) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 84)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 84) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 36) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 91)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 91) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 43) % 48) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 98)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 98) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 2) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 105)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 105) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 9) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 112)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 112) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 16) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 119)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 119) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 23) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 126)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 126) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 30) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 133)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 133) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 37) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 140)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 140) / 48) * 4608)) + (rc_outer_outer * 144)) + (((((int)threadIdx.x) + 44) % 48) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 147)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 147) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 3) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 154)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 154) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 10) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 161)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 161) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 17) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 168)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 168) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 24) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 175)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 175) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 31) * 3)) + rx_outer_outer)];
- kernel_shared[(((int)threadIdx.x) + 182)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 182) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 38) * 3)) + rx_outer_outer)];
- if (((int)threadIdx.x) < 3) {
- kernel_shared[(((int)threadIdx.x) + 189)] = kernel[(((((((int)blockIdx.x) * 18432) + (((((int)threadIdx.x) + 189) / 48) * 4608)) + (rc_outer_outer * 144)) + ((((int)threadIdx.x) + 45) * 3)) + rx_outer_outer)];
+ pad_temp_shared[((int)threadIdx.x)] = (((((7 <= (((int)threadIdx.x) % 63)) && ((((int)threadIdx.x) % 63) < 56)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[(((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 63) * 49)) + rx_outer_outer) + (((int)threadIdx.x) % 63)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 112)] = (((((1 <= (((((int)threadIdx.x) / 7) + 7) % 9)) && ((((((int)threadIdx.x) / 7) + 7) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 112) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 7) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 224) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ pad_temp_shared[(((int)threadIdx.x) + 336)] = (((((1 <= (((((int)threadIdx.x) / 7) + 3) % 9)) && ((((((int)threadIdx.x) / 7) + 3) % 9) < 8)) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 336) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 3) % 9) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ if (((int)threadIdx.x) < 56) {
+ pad_temp_shared[(((int)threadIdx.x) + 448)] = ((((((int)threadIdx.x) < 49) && (1 <= (rx_outer_outer + (((int)threadIdx.x) % 7)))) && ((rx_outer_outer + (((int)threadIdx.x) % 7)) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 448) / 63) * 49)) + (((((int)threadIdx.x) / 7) + 1) * 7)) + rx_outer_outer) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+ }
+ kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 112)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 112) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 16) % 24) * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 224)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 224) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 8) % 24) * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 336)] = kernel[((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3)) + rx_outer_outer) + 64512)];
+ kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 448) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 16) % 24) * 3)) + rx_outer_outer)];
+ kernel_shared[(((int)threadIdx.x) + 560)] = kernel[(((((((int)blockIdx.x) * 147456) + (((((int)threadIdx.x) + 560) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) + 8) % 24) * 3)) + rx_outer_outer)];
+ if (((int)threadIdx.x) < 96) {
+ kernel_shared[(((int)threadIdx.x) + 672)] = kernel[((((((((int)blockIdx.x) * 147456) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((int)threadIdx.x) % 24) * 3)) + rx_outer_outer) + 129024)];
}
__syncthreads();
- for (int rc_outer_inner = 0; rc_outer_inner < 16; ++rc_outer_inner) {
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((rc_outer_inner * 63) + ((int)threadIdx.x))] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((rc_outer_inner * 63) + ((int)threadIdx.x))] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[(rc_outer_inner * 3)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 48)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 1)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 49)]));
- conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 56)] * kernel_shared[((rc_outer_inner * 3) + 2)]));
- conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 56)] * kernel_shared[((rc_outer_inner * 3) + 50)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[((rc_outer_inner * 63) + ((int)threadIdx.x))] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[((rc_outer_inner * 63) + ((int)threadIdx.x))] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[14] = (conv2d_nchw[14] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[21] = (conv2d_nchw[21] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 7)] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[15] = (conv2d_nchw[15] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[22] = (conv2d_nchw[22] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 14)] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[16] = (conv2d_nchw[16] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[23] = (conv2d_nchw[23] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 21)] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[17] = (conv2d_nchw[17] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[24] = (conv2d_nchw[24] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 28)] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[18] = (conv2d_nchw[18] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[25] = (conv2d_nchw[25] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 35)] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[19] = (conv2d_nchw[19] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[26] = (conv2d_nchw[26] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 96)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 42)] * kernel_shared[((rc_outer_inner * 3) + 144)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 97)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 49)] * kernel_shared[((rc_outer_inner * 3) + 145)]));
- conv2d_nchw[20] = (conv2d_nchw[20] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 56)] * kernel_shared[((rc_outer_inner * 3) + 98)]));
- conv2d_nchw[27] = (conv2d_nchw[27] + (pad_temp_shared[(((rc_outer_inner * 63) + ((int)threadIdx.x)) + 56)] * kernel_shared[((rc_outer_inner * 3) + 146)]));
+ for (int rc_outer_inner = 0; rc_outer_inner < 4; ++rc_outer_inner) {
+ for (int ry_outer_inner = 0; ry_outer_inner < 3; ++ry_outer_inner) {
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 63)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 7)] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 7)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 70)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 70)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 14)] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 14)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 77)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 77)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 21)] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 21)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 84)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 84)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 28)] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 28)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 91)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 35)] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 35)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 98)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 98)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 42)] * kernel_shared[((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 42)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 24)]));
+ conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 105)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 3)]));
+ conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[((((rc_outer_inner * 126) + (ry_outer_inner * 7)) + (((int)threadIdx.x) % 7)) + 105)] * kernel_shared[(((((((int)threadIdx.x) / 7) * 48) + (rc_outer_inner * 6)) + ry_outer_inner) + 27)]));
+ }
}
}
}
- for (int i1_inner = 0; i1_inner < 4; ++i1_inner) {
+ for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
for (int i2_inner = 0; i2_inner < 7; ++i2_inner) {
- compute[((((((int)blockIdx.x) * 196) + (i1_inner * 49)) + (i2_inner * 7)) + ((int)threadIdx.x))] = max((conv2d_nchw[((i1_inner * 7) + i2_inner)] + bias[((((int)blockIdx.x) * 4) + i1_inner)]), 0.000000e+00f);
+ compute[(((((((int)blockIdx.x) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + (i2_inner * 7)) + (((int)threadIdx.x) % 7))] = max((conv2d_nchw[((i1_inner * 7) + i2_inner)] + bias[(((((int)blockIdx.x) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
}
}
}
@@ -1440,7 +815,7 @@ In the example below we resume the status and do more 5 trials.</p>
Get devices for measurement successfully!
</pre></div>
</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 39.840 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes 27.183 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/e3e540f3b477c0c52d8eb73e674e8ffd/tune_conv2d_layer_cuda.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">tune_conv2d_layer_cuda.py</span></code></a></p>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
index 2b97959fa..521126505 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -876,7 +876,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 9.7388 9.7416 9.7603 9.7146 0.0188
+ 9.9670 9.9606 10.0306 9.9100 0.0494
</pre></div>
</div>
</div>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
index fd5ae103b..a5c123190 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -895,7 +895,7 @@ so we can read the log file and load the best schedules.</p>
Evaluate inference time cost...
Execution time summary:
mean (ms) median (ms) max (ms) min (ms) std (ms)
- 744.2534 743.2250 747.7378 741.7976 2.5318
+ 755.3439 753.5566 759.6341 752.8411 3.0476
</pre></div>
</div>
</div>
@@ -917,7 +917,7 @@ to learn how to use the RPC Tracker and RPC Server.
To use the RPC Tracker in auto-scheduler, replace the runner in <code class="code docutils literal notranslate"><span class="pre">TuningOptions</span></code>
with <a class="reference internal" href="../../reference/api/python/auto_scheduler.html#tvm.auto_scheduler.RPCRunner" title="tvm.auto_scheduler.RPCRunner"><code class="xref any py py-class docutils literal notranslate"><span class="pre">auto_scheduler.RPCRunner</span></code></a>.</p></li>
</ol>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 18.913 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 20.913 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-network-x86-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/e416b94ca1090b0897c0f6e0df95b911/tune_network_x86.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">tune_network_x86.py</span></code></a></p>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
index 752ccf4a0..572a50511 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -600,217 +600,28 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
- preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), placeholder_5: placeholder_16: Buffer(placeholder_10, float32, [128, 256], []), placeholder_6: placeholder_17: Buffer(placeholder_11, float32, [4916, 16, 1], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_9: placeholder_19: Buffer(placeholder_14, float32, [128, 512], [])} {
- for (i0.outer.i1.outer.fused: int32, 0, 512) "parallel" {
- allocate(compute_4: Pointer(global float32), float32, [128]), storage_scope = global {
- for (nb_j.inner: int32, 0, 2) {
- let cse_var_2: int32 = (nb_j.inner*16)
- let cse_var_1: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
- {
- compute_5: Buffer(compute_4, float32, [128], [])[cse_var_2] = 0f32
- compute_5[(cse_var_2 + 1)] = 0f32
- compute_5[(cse_var_2 + 2)] = 0f32
- compute_5[(cse_var_2 + 3)] = 0f32
- compute_5[(cse_var_2 + 4)] = 0f32
- compute_5[(cse_var_2 + 5)] = 0f32
- compute_5[(cse_var_2 + 6)] = 0f32
- compute_5[(cse_var_2 + 7)] = 0f32
- compute_5[(cse_var_2 + 8)] = 0f32
- compute_5[(cse_var_2 + 9)] = 0f32
- compute_5[(cse_var_2 + 10)] = 0f32
- compute_5[(cse_var_2 + 11)] = 0f32
- compute_5[(cse_var_2 + 12)] = 0f32
- compute_5[(cse_var_2 + 13)] = 0f32
- compute_5[(cse_var_2 + 14)] = 0f32
- compute_5[(cse_var_2 + 15)] = 0f32
- compute_5[(cse_var_2 + 32)] = 0f32
- compute_5[(cse_var_2 + 33)] = 0f32
- compute_5[(cse_var_2 + 34)] = 0f32
- compute_5[(cse_var_2 + 35)] = 0f32
- compute_5[(cse_var_2 + 36)] = 0f32
- compute_5[(cse_var_2 + 37)] = 0f32
- compute_5[(cse_var_2 + 38)] = 0f32
- compute_5[(cse_var_2 + 39)] = 0f32
- compute_5[(cse_var_2 + 40)] = 0f32
- compute_5[(cse_var_2 + 41)] = 0f32
- compute_5[(cse_var_2 + 42)] = 0f32
- compute_5[(cse_var_2 + 43)] = 0f32
- compute_5[(cse_var_2 + 44)] = 0f32
- compute_5[(cse_var_2 + 45)] = 0f32
- compute_5[(cse_var_2 + 46)] = 0f32
- compute_5[(cse_var_2 + 47)] = 0f32
- compute_5[(cse_var_2 + 64)] = 0f32
- compute_5[(cse_var_2 + 65)] = 0f32
- compute_5[(cse_var_2 + 66)] = 0f32
- compute_5[(cse_var_2 + 67)] = 0f32
- compute_5[(cse_var_2 + 68)] = 0f32
- compute_5[(cse_var_2 + 69)] = 0f32
- compute_5[(cse_var_2 + 70)] = 0f32
- compute_5[(cse_var_2 + 71)] = 0f32
- compute_5[(cse_var_2 + 72)] = 0f32
- compute_5[(cse_var_2 + 73)] = 0f32
- compute_5[(cse_var_2 + 74)] = 0f32
- compute_5[(cse_var_2 + 75)] = 0f32
- compute_5[(cse_var_2 + 76)] = 0f32
- compute_5[(cse_var_2 + 77)] = 0f32
- compute_5[(cse_var_2 + 78)] = 0f32
- compute_5[(cse_var_2 + 79)] = 0f32
- compute_5[(cse_var_2 + 96)] = 0f32
- compute_5[(cse_var_2 + 97)] = 0f32
- compute_5[(cse_var_2 + 98)] = 0f32
- compute_5[(cse_var_2 + 99)] = 0f32
- compute_5[(cse_var_2 + 100)] = 0f32
- compute_5[(cse_var_2 + 101)] = 0f32
- compute_5[(cse_var_2 + 102)] = 0f32
- compute_5[(cse_var_2 + 103)] = 0f32
- compute_5[(cse_var_2 + 104)] = 0f32
- compute_5[(cse_var_2 + 105)] = 0f32
- compute_5[(cse_var_2 + 106)] = 0f32
- compute_5[(cse_var_2 + 107)] = 0f32
- compute_5[(cse_var_2 + 108)] = 0f32
- compute_5[(cse_var_2 + 109)] = 0f32
- compute_5[(cse_var_2 + 110)] = 0f32
- compute_5[(cse_var_2 + 111)] = 0f32
- for (elem_idx: int32, 0, (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
- let cse_var_67: int32 = (cse_var_2 + 41)
- let cse_var_66: int32 = (cse_var_2 + 40)
- let cse_var_65: int32 = (cse_var_2 + 4)
- let cse_var_64: int32 = (cse_var_2 + 39)
- let cse_var_63: int32 = (cse_var_2 + 38)
- let cse_var_62: int32 = (cse_var_2 + 37)
- let cse_var_61: int32 = (cse_var_2 + 36)
- let cse_var_60: int32 = (cse_var_2 + 35)
- let cse_var_59: int32 = (cse_var_2 + 34)
- let cse_var_58: int32 = (cse_var_2 + 33)
- let cse_var_57: int32 = (cse_var_2 + 32)
- let cse_var_56: int32 = (cse_var_2 + 3)
- let cse_var_55: int32 = (cse_var_2 + 2)
- let cse_var_54: int32 = (cse_var_2 + 15)
- let cse_var_53: int32 = (cse_var_2 + 14)
- let cse_var_52: int32 = (cse_var_2 + 1)
- let cse_var_51: int32 = (cse_var_2 + 12)
- let cse_var_50: int32 = (cse_var_2 + 111)
- let cse_var_49: int32 = (cse_var_2 + 110)
- let cse_var_48: int32 = (cse_var_2 + 11)
- let cse_var_47: int32 = (cse_var_2 + 109)
- let cse_var_46: int32 = (cse_var_2 + 108)
- let cse_var_45: int32 = (cse_var_2 + 107)
- let cse_var_44: int32 = (cse_var_2 + 106)
- let cse_var_43: int32 = (cse_var_2 + 105)
- let cse_var_42: int32 = (cse_var_2 + 104)
- let cse_var_41: int32 = (cse_var_2 + 103)
- let cse_var_40: int32 = (cse_var_2 + 102)
- let cse_var_39: int32 = (cse_var_2 + 101)
- let cse_var_38: int32 = (cse_var_2 + 100)
- let cse_var_37: int32 = (cse_var_2 + 10)
- let cse_var_36: int32 = (cse_var_2 + 13)
- let cse_var_35: int32 = (elem_idx*16)
- let cse_var_34: int32 = (cse_var_2 + 99)
- let cse_var_33: int32 = (cse_var_2 + 98)
- let cse_var_32: int32 = (cse_var_2 + 97)
- let cse_var_31: int32 = (cse_var_2 + 96)
- let cse_var_30: int32 = (cse_var_2 + 9)
- let cse_var_29: int32 = (cse_var_2 + 8)
- let cse_var_28: int32 = (cse_var_2 + 79)
- let cse_var_27: int32 = (cse_var_2 + 78)
- let cse_var_26: int32 = (cse_var_2 + 77)
- let cse_var_25: int32 = (cse_var_2 + 76)
- let cse_var_24: int32 = (cse_var_2 + 75)
- let cse_var_23: int32 = (cse_var_2 + 74)
- let cse_var_22: int32 = (cse_var_2 + 73)
- let cse_var_21: int32 = (cse_var_2 + 72)
- let cse_var_20: int32 = (cse_var_2 + 71)
- let cse_var_19: int32 = (cse_var_2 + 42)
- let cse_var_18: int32 = (cse_var_2 + 44)
- let cse_var_17: int32 = (cse_var_2 + 45)
- let cse_var_16: int32 = (cse_var_2 + 46)
- let cse_var_15: int32 = (cse_var_2 + 47)
- let cse_var_14: int32 = (cse_var_2 + 5)
- let cse_var_13: int32 = (cse_var_2 + 6)
- let cse_var_12: int32 = (cse_var_2 + 64)
- let cse_var_11: int32 = (cse_var_2 + 65)
- let cse_var_10: int32 = (cse_var_2 + 66)
- let cse_var_9: int32 = (cse_var_2 + 67)
- let cse_var_8: int32 = (cse_var_2 + 68)
- let cse_var_7: int32 = (cse_var_2 + 69)
- let cse_var_6: int32 = (cse_var_2 + 7)
- let cse_var_5: int32 = (cse_var_2 + 43)
- let cse_var_4: int32 = (cse_var_2 + 70)
- let cse_var_3: int32 = (floordiv(i0.outer.i1.outer.fused, 16)*1024)
- {
- compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_35)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_52] = (compute_5[cse_var_52] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 1)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_55] = (compute_5[cse_var_55] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 2)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_56] = (compute_5[cse_var_56] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 3)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_65] = (compute_5[cse_var_65] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 4)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 5)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 6)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 7)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_29] = (compute_5[cse_var_29] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 8)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_30] = (compute_5[cse_var_30] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 9)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_37] = (compute_5[cse_var_37] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 10)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_48] = (compute_5[cse_var_48] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 11)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_51] = (compute_5[cse_var_51] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 12)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_36] = (compute_5[cse_var_36] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 13)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_53] = (compute_5[cse_var_53] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 14)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_54] = (compute_5[cse_var_54] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 15)]*max(placeholder[(cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)])], 0f32)))
- compute_5[cse_var_57] = (compute_5[cse_var_57] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_35)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_58] = (compute_5[cse_var_58] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_59] = (compute_5[cse_var_59] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_60] = (compute_5[cse_var_60] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_61] = (compute_5[cse_var_61] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_62] = (compute_5[cse_var_62] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_63] = (compute_5[cse_var_63] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_64] = (compute_5[cse_var_64] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_66] = (compute_5[cse_var_66] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_67] = (compute_5[cse_var_67] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_19] = (compute_5[cse_var_19] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_18] = (compute_5[cse_var_18] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 256)], 0f32)))
- compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_35)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_20] = (compute_5[cse_var_20] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_21] = (compute_5[cse_var_21] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_22] = (compute_5[cse_var_22] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_23] = (compute_5[cse_var_23] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_24] = (compute_5[cse_var_24] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_25] = (compute_5[cse_var_25] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_26] = (compute_5[cse_var_26] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_27] = (compute_5[cse_var_27] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_28] = (compute_5[cse_var_28] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 512)], 0f32)))
- compute_5[cse_var_31] = (compute_5[cse_var_31] + (placeholder_1[((placeholder_3[cse_var_1]*16) + cse_var_35)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_32] = (compute_5[cse_var_32] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 1)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_33] = (compute_5[cse_var_33] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 2)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_34] = (compute_5[cse_var_34] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 3)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_38] = (compute_5[cse_var_38] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 4)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_39] = (compute_5[cse_var_39] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 5)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_40] = (compute_5[cse_var_40] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 6)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_41] = (compute_5[cse_var_41] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 7)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_42] = (compute_5[cse_var_42] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 8)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_43] = (compute_5[cse_var_43] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 9)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_44] = (compute_5[cse_var_44] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 10)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_45] = (compute_5[cse_var_45] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 11)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_46] = (compute_5[cse_var_46] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 12)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_47] = (compute_5[cse_var_47] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 13)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_49] = (compute_5[cse_var_49] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 14)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
- compute_5[cse_var_50] = (compute_5[cse_var_50] + (placeholder_1[(((placeholder_3[cse_var_1]*16) + cse_var_35) + 15)]*max(placeholder[((cse_var_3 + placeholder_2[(placeholder_3[cse_var_1] + elem_idx)]) + 768)], 0f32)))
+ preflattened_buffer_map = {placeholder_8: placeholder_15: Buffer(placeholder_13, int32, [33], []), placeholder_5: placeholder_16: Buffer(placeholder_10, float32, [128, 256], []), placeholder_9: placeholder_17: Buffer(placeholder_14, float32, [128, 512], []), placeholder_7: placeholder_18: Buffer(placeholder_12, int32, [4916], []), placeholder_6: placeholder_19: Buffer(placeholder_11, float32, [4916, 16, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], [])} {
+ for (i0.outer.i1.outer.fused: int32, 0, 128) "parallel" {
+ allocate(compute_4: Pointer(global float32), float32, [512]), storage_scope = global {
+ for (i.outer.inner: int32, 0, 8) {
+ for (i.inner.init: int32, 0, 4) {
+ for (j.init: int32, 0, 16) {
+ compute_5: Buffer(compute_4, float32, [512], [])[(((i.outer.inner*64) + (i.inner.init*16)) + j.init)] = 0f32
+ }
+ }
+ for (elem_idx: int32, 0, let cse_var_1: int32 = floormod(i0.outer.i1.outer.fused, 32) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
+ for (i.inner: int32, 0, 4) {
+ for (j: int32, 0, 16) {
+ let cse_var_3: int32 = floormod(i0.outer.i1.outer.fused, 32)
+ let cse_var_2: int32 = (((i.outer.inner*64) + (i.inner*16)) + j)
+ compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[((((floordiv(i0.outer.i1.outer.fused, 32)*8192) + (i.outer.inner*1024)) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
}
}
}
}
- for (i0.inner: int32, 0, 4) {
- for (i1.inner: int32, 0, 32) {
- let cse_var_68: int32 = ((((floordiv(i0.outer.i1.outer.fused, 16)*2048) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32)) + i1.inner)
- compute[cse_var_68] = max((compute_5[((i0.inner*32) + i1.inner)] + placeholder_4[cse_var_68]), 0f32)
- }
+ for (i0.inner: int32, 0, 32) {
+ let cse_var_4: int32 = (((floordiv(i0.outer.i1.outer.fused, 32)*16384) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 32)*16))
+ compute[ramp(cse_var_4, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_4, 1, 16)]), broadcast(0f32, 16))
}
}
}
@@ -849,7 +660,7 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 2.587 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.474 ms
</pre></div>
</div>
<div class="admonition note">
diff --git a/docs/how_to/tune_with_autotvm/sg_execution_times.html b/docs/how_to/tune_with_autotvm/sg_execution_times.html
index 1ab792ff7..e28f359c5 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -300,13 +300,13 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-tune-with-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:44.049</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:44.195</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:43.190</strong>: <a class="reference internal" href="tune_conv2d_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-conv2d-cuda-py"><span class="std std-ref">Tuning High Performance Convolution on NVIDIA GPUs</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_cuda.py</span></code>)</p></li>
-<li><p><strong>00:00.224</strong>: <a class="reference internal" href="tune_relay_x86.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-x86-py"><span class="std std-ref">Auto-tuning a Convolutional Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_x86.py</span></code>)</p></li>
-<li><p><strong>00:00.213</strong>: <a class="reference internal" href="tune_relay_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-cuda-py"><span class="std std-ref">Auto-tuning a Convolutional Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_cuda.py</span></code>)</p></li>
-<li><p><strong>00:00.211</strong>: <a class="reference internal" href="tune_relay_mobile_gpu.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-mobile-gpu-py"><span class="std std-ref">Auto-tuning a Convolutional Network for Mobile GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_mobile_gpu.py</span></code>)</p></li>
-<li><p><strong>00:00.211</strong>: <a class="reference internal" href="tune_relay_arm.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-arm-py"><span class="std std-ref">Auto-tuning a Convolutional Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_arm.py</span></code>)</p></li>
+<li><p><strong>00:43.373</strong>: <a class="reference internal" href="tune_conv2d_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-conv2d-cuda-py"><span class="std std-ref">Tuning High Performance Convolution on NVIDIA GPUs</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_cuda.py</span></code>)</p></li>
+<li><p><strong>00:00.217</strong>: <a class="reference internal" href="tune_relay_x86.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-x86-py"><span class="std std-ref">Auto-tuning a Convolutional Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_x86.py</span></code>)</p></li>
+<li><p><strong>00:00.205</strong>: <a class="reference internal" href="tune_relay_mobile_gpu.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-mobile-gpu-py"><span class="std std-ref">Auto-tuning a Convolutional Network for Mobile GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_mobile_gpu.py</span></code>)</p></li>
+<li><p><strong>00:00.200</strong>: <a class="reference internal" href="tune_relay_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-cuda-py"><span class="std std-ref">Auto-tuning a Convolutional Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_cuda.py</span></code>)</p></li>
+<li><p><strong>00:00.200</strong>: <a class="reference internal" href="tune_relay_arm.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-arm-py"><span class="std std-ref">Auto-tuning a Convolutional Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_arm.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
index 1ccb658b3..5feb8f9e3 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -1142,8 +1142,8 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 4, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2885496
-No: 6 GFLOPS: 42.26/42.26 result: MeasureResult(costs=(0.0054775518947368425,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5810227394104004, timestamp=1650670409.569581) [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
-No: 7 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+No: 6 GFLOPS: 102.59/102.59 result: MeasureResult(costs=(0.0022566282291666667,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5949983596801758, timestamp=1650690027.666691) [('tile_f', [-1, 1, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3754080
+No: 7 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1266,7 +1266,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 16, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 256, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6225319
-No: 8 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+No: 8 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1389,7 +1389,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 32]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,943546
-No: 9 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+No: 9 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1512,7 +1512,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 4, 16, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2868708
-No: 10 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+No: 10 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 142, in build
res = future.result()
File "/usr/lib/python3.7/concurrent/futures/_base.py", line 435, in result
@@ -1530,7 +1530,7 @@ No: 10 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
TimeoutError
[('tile_f', [-1, 32, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4691833
-No: 11 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+No: 11 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1653,7 +1653,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 2, 64]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1042124
-No: 12 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+No: 12 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1776,7 +1776,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 32, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10013405
-No: 13 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+No: 13 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -1899,7 +1899,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 8, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6732082
-No: 14 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+No: 14 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2022,7 +2022,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 4, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7536735
-No: 15 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+No: 15 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2145,7 +2145,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,482121
-No: 16 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+No: 16 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2268,7 +2268,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 2, 1, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2824525
-No: 17 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+No: 17 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2391,7 +2391,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 64, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4559286
-No: 18 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+No: 18 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 571, in __call__
func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 523, in _build_func_common
@@ -2514,7 +2514,7 @@ Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 854, in verify_pass
raise InstantiationError("Skipped because of invalid gpu kernel")
tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel [('tile_f', [-1, 1, 32, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9677544
-No: 19 GFLOPS: 0.00/42.26 result: Traceback (most recent call last):
+No: 19 GFLOPS: 0.00/102.59 result: Traceback (most recent call last):
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 721, in __call__
yield remote, remote.load_module(os.path.split(build_result.filename)[1])
File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 685, in run_through_rpc
@@ -2602,7 +2602,7 @@ tvm._ffi.base.TVMError: Traceback (most recent call last):
15: _PyEval_EvalFrameDefault
14: 0x0000000000537c30
13: _PyObject_FastCallKeywords
- 12: 0x00007fce6fd74fa2
+ 12: 0x00007f48339cefa2
11: _ctypes_callproc
10: ffi_call
9: ffi_call_unix64
@@ -2667,7 +2667,7 @@ Traceback (most recent call last):
21: _PyFunction_FastCallKeywords
20: _PyEval_EvalFrameDefault
19: _PyFunction_FastCall [('tile_f', [-1, 8, 2, 16]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6390073
-No: 20 GFLOPS: 144.55/144.55 result: MeasureResult(costs=(0.0016015395100000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3942515850067139, timestamp=1650670435.7232537) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
+No: 20 GFLOPS: 142.38/142.38 result: MeasureResult(costs=(0.0016259662,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.420372486114502, timestamp=1650690054.0953064) [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
</pre></div>
</div>
<p>Finally we can inspect the best config from log file, check correctness,
@@ -2706,7 +2706,7 @@ and measure running time.</p>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Best config:
[('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9881539
-Time cost of this operator: 0.001962
+Time cost of this operator: 0.001982
</pre></div>
</div>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autotvm-tune-conv2d-cuda-py">
diff --git a/docs/how_to/work_with_microtvm/micro_autotune.html b/docs/how_to/work_with_microtvm/micro_autotune.html
index 1665a7044..12d150c80 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -553,10 +553,10 @@ the tuned operator.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build without Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs
--------- --- -------- ------- ----- ------ -------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 305.2 98.718 (1, 2, 10, 10, 3) 2 1
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.018 0.976 (1, 6, 10, 10) 1 1
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.945 0.306 (1, 1, 10, 10, 3) 1 1
-Total_time - 309.163 - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 312.8 98.738 (1, 2, 10, 10, 3) 2 1
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 3.073 0.97 (1, 6, 10, 10) 1 1
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.925 0.292 (1, 1, 10, 10, 3) 1 1
+Total_time - 316.798 - - - -
</pre></div>
</div>
</div>
@@ -608,10 +608,10 @@ Total_time -
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build with Autotuning ##########
Node Name Ops Time(us) Time(%) Shape Inputs Outputs
--------- --- -------- ------- ----- ------ -------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 150.5 98.203 (1, 6, 10, 10, 1) 2 1
-tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.805 1.178 (1, 6, 10, 10) 1 1
-tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.948 0.619 (1, 1, 10, 10, 3) 1 1
-Total_time - 153.253 - - - -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc tvmgen_default_fused_nn_contrib_conv2d_NCHWc 81.0 96.817 (1, 6, 10, 10, 1) 2 1
+tvmgen_default_fused_layout_transform_1 tvmgen_default_fused_layout_transform_1 1.728 2.065 (1, 6, 10, 10) 1 1
+tvmgen_default_fused_layout_transform tvmgen_default_fused_layout_transform 0.936 1.118 (1, 1, 10, 10, 3) 1 1
+Total_time - 83.663 - - - -
</pre></div>
</div>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-autotune-py">
diff --git a/docs/how_to/work_with_microtvm/sg_execution_times.html b/docs/how_to/work_with_microtvm/sg_execution_times.html
index 939686578..c97d5519e 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -300,13 +300,13 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-microtvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:43.963</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>00:44.000</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:39.912</strong>: <a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">Autotuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></li>
-<li><p><strong>00:03.462</strong>: <a class="reference internal" href="micro_tflite.html#sphx-glr-how-to-work-with-microtvm-micro-tflite-py"><span class="std std-ref">microTVM with TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></li>
-<li><p><strong>00:00.199</strong>: <a class="reference internal" href="micro_tvmc.html#sphx-glr-how-to-work-with-microtvm-micro-tvmc-py"><span class="std std-ref">Executing a Tiny Model with TVMC Micro</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tvmc.py</span></code>)</p></li>
-<li><p><strong>00:00.197</strong>: <a class="reference internal" href="micro_reference_vm.html#sphx-glr-how-to-work-with-microtvm-micro-reference-vm-py"><span class="std std-ref">microTVM Reference Virtual Machines</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_reference_vm.py</span></code>)</p></li>
-<li><p><strong>00:00.193</strong>: <a class="reference internal" href="micro_ethosu.html#sphx-glr-how-to-work-with-microtvm-micro-ethosu-py"><span class="std std-ref">Running TVM on bare metal Arm(R) Cortex(R)-M55 CPU and Ethos(TM)-U55 NPU</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></li>
+<li><p><strong>00:39.918</strong>: <a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">Autotuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></li>
+<li><p><strong>00:03.494</strong>: <a class="reference internal" href="micro_tflite.html#sphx-glr-how-to-work-with-microtvm-micro-tflite-py"><span class="std std-ref">microTVM with TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></li>
+<li><p><strong>00:00.208</strong>: <a class="reference internal" href="micro_reference_vm.html#sphx-glr-how-to-work-with-microtvm-micro-reference-vm-py"><span class="std std-ref">microTVM Reference Virtual Machines</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_reference_vm.py</span></code>)</p></li>
+<li><p><strong>00:00.191</strong>: <a class="reference internal" href="micro_ethosu.html#sphx-glr-how-to-work-with-microtvm-micro-ethosu-py"><span class="std std-ref">Running TVM on bare metal Arm(R) Cortex(R)-M55 CPU and Ethos(TM)-U55 NPU</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></li>
+<li><p><strong>00:00.190</strong>: <a class="reference internal" href="micro_tvmc.html#sphx-glr-how-to-work-with-microtvm-micro-tvmc-py"><span class="std std-ref">Executing a Tiny Model with TVMC Micro</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tvmc.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/work_with_relay/sg_execution_times.html b/docs/how_to/work_with_relay/sg_execution_times.html
index c88282d9f..1a9ac2fa3 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -300,10 +300,10 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-relay-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:09.266</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:09.154</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:07.223</strong>: <a class="reference internal" href="using_external_lib.html#sphx-glr-how-to-work-with-relay-using-external-lib-py"><span class="std std-ref">Using External Libraries in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_external_lib.py</span></code>)</p></li>
-<li><p><strong>00:01.830</strong>: <a class="reference internal" href="build_gcn.html#sphx-glr-how-to-work-with-relay-build-gcn-py"><span class="std std-ref">Building a Graph Convolutional Network</span></a> (<code class="docutils literal notranslate"><span class="pre">build_gcn.py</span></code>)</p></li>
+<li><p><strong>00:07.437</strong>: <a class="reference internal" href="using_external_lib.html#sphx-glr-how-to-work-with-relay-using-external-lib-py"><span class="std std-ref">Using External Libraries in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_external_lib.py</span></code>)</p></li>
+<li><p><strong>00:01.503</strong>: <a class="reference internal" href="build_gcn.html#sphx-glr-how-to-work-with-relay-build-gcn-py"><span class="std std-ref">Building a Graph Convolutional Network</span></a> (<code class="docutils literal notranslate"><span class="pre">build_gcn.py</span></code>)</p></li>
<li><p><strong>00:00.214</strong>: <a class="reference internal" href="using_relay_viz.html#sphx-glr-how-to-work-with-relay-using-relay-viz-py"><span class="std std-ref">Use Relay Visualizer to Visualize Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_relay_viz.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/work_with_schedules/sg_execution_times.html b/docs/how_to/work_with_schedules/sg_execution_times.html
index 66b7a1865..e3c17b1d3 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -300,16 +300,16 @@
<div class="section" id="computation-times">
<span id="sphx-glr-how-to-work-with-schedules-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:05.515</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:05.549</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:02.014</strong>: <a class="reference internal" href="intrin_math.html#sphx-glr-how-to-work-with-schedules-intrin-math-py"><span class="std std-ref">Intrinsics and Math Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">intrin_math.py</span></code>)</p></li>
-<li><p><strong>00:01.101</strong>: <a class="reference internal" href="tensorize.html#sphx-glr-how-to-work-with-schedules-tensorize-py"><span class="std std-ref">Use Tensorize to Leverage Hardware Intrinsics</span></a> (<code class="docutils literal notranslate"><span class="pre">tensorize.py</span></code>)</p></li>
-<li><p><strong>00:00.712</strong>: <a class="reference internal" href="reduction.html#sphx-glr-how-to-work-with-schedules-reduction-py"><span class="std std-ref">Reduction</span></a> (<code class="docutils literal notranslate"><span class="pre">reduction.py</span></code>)</p></li>
-<li><p><strong>00:00.692</strong>: <a class="reference internal" href="scan.html#sphx-glr-how-to-work-with-schedules-scan-py"><span class="std std-ref">Scan and Recurrent Kernel</span></a> (<code class="docutils literal notranslate"><span class="pre">scan.py</span></code>)</p></li>
-<li><p><strong>00:00.304</strong>: <a class="reference internal" href="extern_op.html#sphx-glr-how-to-work-with-schedules-extern-op-py"><span class="std std-ref">External Tensor Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">extern_op.py</span></code>)</p></li>
-<li><p><strong>00:00.237</strong>: <a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></li>
-<li><p><strong>00:00.236</strong>: <a class="reference internal" href="tedd.html#sphx-glr-how-to-work-with-schedules-tedd-py"><span class="std std-ref">Use Tensor Expression Debug Display (TEDD) for Visualization</span></a> (<code class="docutils literal notranslate"><span class="pre">tedd.py</span></code>)</p></li>
-<li><p><strong>00:00.218</strong>: <a class="reference internal" href="tuple_inputs.html#sphx-glr-how-to-work-with-schedules-tuple-inputs-py"><span class="std std-ref">Compute and Reduce with Tuple Inputs</span></a> (<code class="docutils literal notranslate"><span class="pre">tuple_inputs.py</span></code>)</p></li>
+<li><p><strong>00:02.037</strong>: <a class="reference internal" href="intrin_math.html#sphx-glr-how-to-work-with-schedules-intrin-math-py"><span class="std std-ref">Intrinsics and Math Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">intrin_math.py</span></code>)</p></li>
+<li><p><strong>00:01.145</strong>: <a class="reference internal" href="tensorize.html#sphx-glr-how-to-work-with-schedules-tensorize-py"><span class="std std-ref">Use Tensorize to Leverage Hardware Intrinsics</span></a> (<code class="docutils literal notranslate"><span class="pre">tensorize.py</span></code>)</p></li>
+<li><p><strong>00:00.713</strong>: <a class="reference internal" href="reduction.html#sphx-glr-how-to-work-with-schedules-reduction-py"><span class="std std-ref">Reduction</span></a> (<code class="docutils literal notranslate"><span class="pre">reduction.py</span></code>)</p></li>
+<li><p><strong>00:00.690</strong>: <a class="reference internal" href="scan.html#sphx-glr-how-to-work-with-schedules-scan-py"><span class="std std-ref">Scan and Recurrent Kernel</span></a> (<code class="docutils literal notranslate"><span class="pre">scan.py</span></code>)</p></li>
+<li><p><strong>00:00.298</strong>: <a class="reference internal" href="extern_op.html#sphx-glr-how-to-work-with-schedules-extern-op-py"><span class="std std-ref">External Tensor Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">extern_op.py</span></code>)</p></li>
+<li><p><strong>00:00.227</strong>: <a class="reference internal" href="tedd.html#sphx-glr-how-to-work-with-schedules-tedd-py"><span class="std std-ref">Use Tensor Expression Debug Display (TEDD) for Visualization</span></a> (<code class="docutils literal notranslate"><span class="pre">tedd.py</span></code>)</p></li>
+<li><p><strong>00:00.224</strong>: <a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></li>
+<li><p><strong>00:00.217</strong>: <a class="reference internal" href="tuple_inputs.html#sphx-glr-how-to-work-with-schedules-tuple-inputs-py"><span class="std std-ref">Compute and Reduce with Tuple Inputs</span></a> (<code class="docutils literal notranslate"><span class="pre">tuple_inputs.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/how_to/work_with_schedules/tensorize.html b/docs/how_to/work_with_schedules/tensorize.html
index 0319237bd..f2b86b104 100644
--- a/docs/how_to/work_with_schedules/tensorize.html
+++ b/docs/how_to/work_with_schedules/tensorize.html
@@ -552,7 +552,7 @@ The importing needs to happen before the tensorized GEMV being executed.</p>
C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
buffer_map = {A_1: A, B_1: B, C_1: C}
preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
- attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpqgpys2ao/input0.cc'\nsource_filename = \"/tmp/tmpqgpys2ao/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = allo [...]
+ attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmp6yusbrem/input0.cc'\nsource_filename = \"/tmp/tmp6yusbrem/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n %7 = allo [...]
for (i, 0, 1024) {
for (j.outer: int32, 0, 32) {
@tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
diff --git a/docs/objects.inv b/docs/objects.inv
index 327ebcd79..6b67cf246 100644
Binary files a/docs/objects.inv and b/docs/objects.inv differ
diff --git a/docs/reference/api/python/auto_scheduler.html b/docs/reference/api/python/auto_scheduler.html
index d1f49ab55..a9484c274 100644
--- a/docs/reference/api/python/auto_scheduler.html
+++ b/docs/reference/api/python/auto_scheduler.html
@@ -1713,7 +1713,7 @@ Can be the a function or the function name.</p></li>
<dl class="py function">
<dt class="sig sig-object py" id="tvm.auto_scheduler.auto_schedule">
-<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
+<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
<dd><p>THIS API IS DEPRECATED.</p>
<p>Run auto scheduling search for a task.</p>
<dl class="field-list simple">
@@ -1750,7 +1750,7 @@ the initial naive schedule (state).</p>
<dl class="py class">
<dt class="sig sig-object py" id="tvm.auto_scheduler.SketchPolicy">
-<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
+<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
<dd><p>The search policy that searches in a hierarchical search space defined by sketches.
The policy randomly samples programs from the space defined by sketches and use evolutionary
search to fine-tune them.</p>
diff --git a/docs/reference/api/typedoc/classes/bytestreamreader.html b/docs/reference/api/typedoc/classes/bytestreamreader.html
index 848659b29..9f39b7893 100644
--- a/docs/reference/api/typedoc/classes/bytestreamreader.html
+++ b/docs/reference/api/typedoc/classes/bytestreamreader.html
@@ -119,7 +119,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -141,7 +141,7 @@
<div class="tsd-signature tsd-kind-icon">bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Uint8Array</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
</ul>
</aside>
</section>
@@ -151,7 +151,7 @@
<div class="tsd-signature tsd-kind-icon">offset<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 0</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L42">rpc_server.ts:42</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L42">rpc_server.ts:42</a></li>
</ul>
</aside>
</section>
@@ -168,7 +168,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L63">rpc_server.ts:63</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L63">rpc_server.ts:63</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">Uint8Array</span></h4>
@@ -185,7 +185,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L49">rpc_server.ts:49</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L49">rpc_server.ts:49</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -202,7 +202,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L57">rpc_server.ts:57</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L57">rpc_server.ts:57</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/cachedcallstack.html b/docs/reference/api/typedoc/classes/cachedcallstack.html
index c148e24b9..79c6b638c 100644
--- a/docs/reference/api/typedoc/classes/cachedcallstack.html
+++ b/docs/reference/api/typedoc/classes/cachedcallstack.html
@@ -144,7 +144,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L223">memory.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L223">memory.ts:223</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -172,7 +172,7 @@
<div class="tsd-signature tsd-kind-icon">temp<wbr>Args<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><a href="../interfaces/disposable.html" class="tsd-signature-type">Disposable</a><span class="tsd-signature-symbol">></span><span class="tsd-signature-symbol"> = []</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L208">memory.ts:208</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L208">memory.ts:208</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -194,7 +194,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L312">memory.ts:312</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L312">memory.ts:312</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -226,7 +226,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L284">memory.ts:284</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L284">memory.ts:284</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -262,7 +262,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L388">memory.ts:388</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L388">memory.ts:388</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -300,7 +300,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L376">memory.ts:376</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L376">memory.ts:376</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -340,7 +340,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L267">memory.ts:267</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L267">memory.ts:267</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -373,7 +373,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L243">memory.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L243">memory.ts:243</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -390,7 +390,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L321">memory.ts:321</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L321">memory.ts:321</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -422,7 +422,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L252">memory.ts:252</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L252">memory.ts:252</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -444,7 +444,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L359">memory.ts:359</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L359">memory.ts:359</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -470,7 +470,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L342">memory.ts:342</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L342">memory.ts:342</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -496,7 +496,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L350">memory.ts:350</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L350">memory.ts:350</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -522,7 +522,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L326">memory.ts:326</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L326">memory.ts:326</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -548,7 +548,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L363">memory.ts:363</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L363">memory.ts:363</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -574,7 +574,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L346">memory.ts:346</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L346">memory.ts:346</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -600,7 +600,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L334">memory.ts:334</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L334">memory.ts:334</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/dldatatype.html b/docs/reference/api/typedoc/classes/dldatatype.html
index 475465e12..629532ede 100644
--- a/docs/reference/api/typedoc/classes/dldatatype.html
+++ b/docs/reference/api/typedoc/classes/dldatatype.html
@@ -119,7 +119,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L262">runtime.ts:262</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L262">runtime.ts:262</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -147,7 +147,7 @@
<div class="tsd-signature tsd-kind-icon">bits<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L260">runtime.ts:260</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L260">runtime.ts:260</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -162,7 +162,7 @@
<div class="tsd-signature tsd-kind-icon">code<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L258">runtime.ts:258</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L258">runtime.ts:258</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -177,7 +177,7 @@
<div class="tsd-signature tsd-kind-icon">lanes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L262">runtime.ts:262</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L262">runtime.ts:262</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -199,7 +199,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L279">runtime.ts:279</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L279">runtime.ts:279</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -216,7 +216,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L270">runtime.ts:270</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L270">runtime.ts:270</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/dldevice.html b/docs/reference/api/typedoc/classes/dldevice.html
index b2bfb496b..382b466cf 100644
--- a/docs/reference/api/typedoc/classes/dldevice.html
+++ b/docs/reference/api/typedoc/classes/dldevice.html
@@ -118,7 +118,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L202">runtime.ts:202</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L202">runtime.ts:202</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -146,7 +146,7 @@
<div class="tsd-signature tsd-kind-icon">device<wbr>Id<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L200">runtime.ts:200</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L200">runtime.ts:200</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -161,7 +161,7 @@
<div class="tsd-signature tsd-kind-icon">device<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L198">runtime.ts:198</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L198">runtime.ts:198</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -183,7 +183,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L223">runtime.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L223">runtime.ts:223</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -205,7 +205,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L230">runtime.ts:230</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L230">runtime.ts:230</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/environment.html b/docs/reference/api/typedoc/classes/environment.html
index 398bd30ba..4fa2c09c3 100644
--- a/docs/reference/api/typedoc/classes/environment.html
+++ b/docs/reference/api/typedoc/classes/environment.html
@@ -125,7 +125,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/environment.ts#L86">environment.ts:86</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/environment.ts#L86">environment.ts:86</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -169,7 +169,7 @@
<aside class="tsd-sources">
<p>Implementation of <a href="../interfaces/libraryprovider.html">LibraryProvider</a>.<a href="../interfaces/libraryprovider.html#imports">imports</a></p>
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/environment.ts#L70">environment.ts:70</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/environment.ts#L70">environment.ts:70</a></li>
</ul>
</aside>
</section>
@@ -179,7 +179,7 @@
<div class="tsd-signature tsd-kind-icon">logger<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>msg<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/environment.ts#L69">environment.ts:69</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/environment.ts#L69">environment.ts:69</a></li>
</ul>
</aside>
<div class="tsd-type-declaration">
@@ -210,7 +210,7 @@
<div class="tsd-signature tsd-kind-icon">packedCFunc<wbr>Table<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">ctypes.FTVMWasmPackedCFunc</span><span class="tsd-signature-symbol"> | </span><span class="tsd-signature-type">undefined</span><span class="tsd-signature-symbol">></span><span class="tsd-signature-symbol"> = [undefined,]</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/environment.ts#L78">environment.ts:78</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/environment.ts#L78">environment.ts:78</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -228,7 +228,7 @@
<div class="tsd-signature tsd-kind-icon">packedCFunc<wbr>Table<wbr>Free<wbr>Id<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">number</span><span class="tsd-signature-symbol">></span><span class="tsd-signature-symbol"> = []</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/environment.ts#L84">environment.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/environment.ts#L84">environment.ts:84</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -250,7 +250,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/environment.ts#L105">environment.ts:105</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/environment.ts#L105">environment.ts:105</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/ffilibrary.html b/docs/reference/api/typedoc/classes/ffilibrary.html
index 16ef45f7c..808a39a31 100644
--- a/docs/reference/api/typedoc/classes/ffilibrary.html
+++ b/docs/reference/api/typedoc/classes/ffilibrary.html
@@ -131,7 +131,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L49">runtime.ts:49</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L49">runtime.ts:49</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -156,7 +156,7 @@
<div class="tsd-signature tsd-kind-icon">exports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">Function</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L46">runtime.ts:46</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L46">runtime.ts:46</a></li>
</ul>
</aside>
</section>
@@ -166,7 +166,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L45">runtime.ts:45</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L45">runtime.ts:45</a></li>
</ul>
</aside>
</section>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">wasm32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">boolean</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L44">runtime.ts:44</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L44">runtime.ts:44</a></li>
</ul>
</aside>
</section>
@@ -186,7 +186,7 @@
<div class="tsd-signature tsd-kind-icon">webGPUContext<span class="tsd-signature-symbol">:</span> <a href="webgpucontext.html" class="tsd-signature-type">WebGPUContext</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L47">runtime.ts:47</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L47">runtime.ts:47</a></li>
</ul>
</aside>
</section>
@@ -203,7 +203,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L76">runtime.ts:76</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L76">runtime.ts:76</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -226,7 +226,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L66">runtime.ts:66</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L66">runtime.ts:66</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -243,7 +243,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L84">runtime.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L84">runtime.ts:84</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <a href="cachedcallstack.html" class="tsd-signature-type">CachedCallStack</a></h4>
@@ -260,7 +260,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L95">runtime.ts:95</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L95">runtime.ts:95</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -283,7 +283,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L72">runtime.ts:72</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L72">runtime.ts:72</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/graphexecutor.html b/docs/reference/api/typedoc/classes/graphexecutor.html
index 64b518147..71fa1e9f4 100644
--- a/docs/reference/api/typedoc/classes/graphexecutor.html
+++ b/docs/reference/api/typedoc/classes/graphexecutor.html
@@ -130,7 +130,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L583">runtime.ts:583</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L583">runtime.ts:583</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -162,7 +162,7 @@
<div class="tsd-signature tsd-kind-icon">module<span class="tsd-signature-symbol">:</span> <a href="module.html" class="tsd-signature-type">Module</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L579">runtime.ts:579</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L579">runtime.ts:579</a></li>
</ul>
</aside>
</section>
@@ -179,7 +179,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L654">runtime.ts:654</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L654">runtime.ts:654</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L597">runtime.ts:597</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L597">runtime.ts:597</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -241,7 +241,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L631">runtime.ts:631</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L631">runtime.ts:631</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -279,7 +279,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L644">runtime.ts:644</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L644">runtime.ts:644</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -310,7 +310,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L621">runtime.ts:621</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L621">runtime.ts:621</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -332,7 +332,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L609">runtime.ts:609</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L609">runtime.ts:609</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/instance.html b/docs/reference/api/typedoc/classes/instance.html
index 72bb77934..c3b1be0d2 100644
--- a/docs/reference/api/typedoc/classes/instance.html
+++ b/docs/reference/api/typedoc/classes/instance.html
@@ -139,7 +139,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L692">runtime.ts:692</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L692">runtime.ts:692</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -202,7 +202,7 @@
<div class="tsd-signature tsd-kind-icon">exports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">Function</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L684">runtime.ts:684</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L684">runtime.ts:684</a></li>
</ul>
</aside>
</section>
@@ -212,7 +212,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L683">runtime.ts:683</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L683">runtime.ts:683</a></li>
</ul>
</aside>
</section>
@@ -229,7 +229,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L932">runtime.ts:932</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L932">runtime.ts:932</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -260,7 +260,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L994">runtime.ts:994</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L994">runtime.ts:994</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -303,7 +303,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L924">runtime.ts:924</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L924">runtime.ts:924</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -341,7 +341,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L732">runtime.ts:732</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L732">runtime.ts:732</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -358,7 +358,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L952">runtime.ts:952</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L952">runtime.ts:952</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -402,7 +402,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L816">runtime.ts:816</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L816">runtime.ts:816</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -434,7 +434,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L1033">runtime.ts:1033</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L1033">runtime.ts:1033</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -465,7 +465,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L846">runtime.ts:846</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L846">runtime.ts:846</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -497,7 +497,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L750">runtime.ts:750</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L750">runtime.ts:750</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -520,7 +520,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L1013">runtime.ts:1013</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L1013">runtime.ts:1013</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -568,7 +568,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L789">runtime.ts:789</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L789">runtime.ts:789</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -608,7 +608,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L914">runtime.ts:914</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L914">runtime.ts:914</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -646,7 +646,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L1134">runtime.ts:1134</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L1134">runtime.ts:1134</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -698,7 +698,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L740">runtime.ts:740</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L740">runtime.ts:740</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -722,7 +722,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L868">runtime.ts:868</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L868">runtime.ts:868</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -754,7 +754,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L857">runtime.ts:857</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L857">runtime.ts:857</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -786,7 +786,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L940">runtime.ts:940</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L940">runtime.ts:940</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/memory.html b/docs/reference/api/typedoc/classes/memory.html
index 3718939b7..4cd7026fc 100644
--- a/docs/reference/api/typedoc/classes/memory.html
+++ b/docs/reference/api/typedoc/classes/memory.html
@@ -130,7 +130,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L40">memory.ts:40</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L40">memory.ts:40</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -152,7 +152,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Memory</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L32">memory.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L32">memory.ts:32</a></li>
</ul>
</aside>
</section>
@@ -162,7 +162,7 @@
<div class="tsd-signature tsd-kind-icon">wasm32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">boolean</span><span class="tsd-signature-symbol"> = true</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L33">memory.ts:33</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L33">memory.ts:33</a></li>
</ul>
</aside>
</section>
@@ -179,7 +179,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L154">memory.ts:154</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L154">memory.ts:154</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -210,7 +210,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L90">memory.ts:90</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L90">memory.ts:90</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -233,7 +233,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L97">memory.ts:97</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L97">memory.ts:97</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -256,7 +256,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L74">memory.ts:74</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L74">memory.ts:74</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -279,7 +279,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L81">memory.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L81">memory.ts:81</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -302,7 +302,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L104">memory.ts:104</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L104">memory.ts:104</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -325,7 +325,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L132">memory.ts:132</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L132">memory.ts:132</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -362,7 +362,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L145">memory.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L145">memory.ts:145</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -393,7 +393,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L60">memory.ts:60</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L60">memory.ts:60</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -416,7 +416,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L67">memory.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L67">memory.ts:67</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -439,7 +439,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L53">memory.ts:53</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L53">memory.ts:53</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -462,7 +462,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L114">memory.ts:114</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L114">memory.ts:114</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -485,7 +485,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L124">memory.ts:124</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L124">memory.ts:124</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -502,7 +502,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/memory.ts#L175">memory.ts:175</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/memory.ts#L175">memory.ts:175</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/module.html b/docs/reference/api/typedoc/classes/module.html
index 6972eade1..a49e344b7 100644
--- a/docs/reference/api/typedoc/classes/module.html
+++ b/docs/reference/api/typedoc/classes/module.html
@@ -124,7 +124,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L504">runtime.ts:504</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L504">runtime.ts:504</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -170,7 +170,7 @@
<div class="tsd-signature tsd-kind-icon">handle<span class="tsd-signature-symbol">:</span> <a href="../index.html#pointer" class="tsd-signature-type">Pointer</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L502">runtime.ts:502</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L502">runtime.ts:502</a></li>
</ul>
</aside>
</section>
@@ -187,7 +187,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L516">runtime.ts:516</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L516">runtime.ts:516</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -204,7 +204,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L530">runtime.ts:530</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L530">runtime.ts:530</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -236,7 +236,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L561">runtime.ts:561</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L561">runtime.ts:561</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/ndarray.html b/docs/reference/api/typedoc/classes/ndarray.html
index bc2b42c29..4fd65beb5 100644
--- a/docs/reference/api/typedoc/classes/ndarray.html
+++ b/docs/reference/api/typedoc/classes/ndarray.html
@@ -130,7 +130,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L304">runtime.ts:304</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L304">runtime.ts:304</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -158,7 +158,7 @@
<div class="tsd-signature tsd-kind-icon">device<span class="tsd-signature-symbol">:</span> <a href="dldevice.html" class="tsd-signature-type">DLDevice</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L297">runtime.ts:297</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L297">runtime.ts:297</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -173,7 +173,7 @@
<div class="tsd-signature tsd-kind-icon">dtype<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L293">runtime.ts:293</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L293">runtime.ts:293</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -188,7 +188,7 @@
<div class="tsd-signature tsd-kind-icon">handle<span class="tsd-signature-symbol">:</span> <a href="../index.html#pointer" class="tsd-signature-type">Pointer</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L289">runtime.ts:289</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L289">runtime.ts:289</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -203,7 +203,7 @@
<div class="tsd-signature tsd-kind-icon">ndim<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L291">runtime.ts:291</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L291">runtime.ts:291</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -218,7 +218,7 @@
<div class="tsd-signature tsd-kind-icon">shape<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">number</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L295">runtime.ts:295</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L295">runtime.ts:295</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -240,7 +240,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L370">runtime.ts:370</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L370">runtime.ts:370</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -273,7 +273,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L414">runtime.ts:414</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L414">runtime.ts:414</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -305,7 +305,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L355">runtime.ts:355</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L355">runtime.ts:355</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -322,7 +322,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L474">runtime.ts:474</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L474">runtime.ts:474</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -346,7 +346,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L443">runtime.ts:443</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L443">runtime.ts:443</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/packedfunccell.html b/docs/reference/api/typedoc/classes/packedfunccell.html
index f4e43d3a2..0a96b4fe3 100644
--- a/docs/reference/api/typedoc/classes/packedfunccell.html
+++ b/docs/reference/api/typedoc/classes/packedfunccell.html
@@ -122,7 +122,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L158">runtime.ts:158</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L158">runtime.ts:158</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -147,7 +147,7 @@
<div class="tsd-signature tsd-kind-icon">handle<span class="tsd-signature-symbol">:</span> <a href="../index.html#pointer" class="tsd-signature-type">Pointer</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L157">runtime.ts:157</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L157">runtime.ts:157</a></li>
</ul>
</aside>
</section>
@@ -164,7 +164,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L165">runtime.ts:165</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L165">runtime.ts:165</a></li>
</ul>
</aside>
<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
diff --git a/docs/reference/api/typedoc/classes/rpcserver.html b/docs/reference/api/typedoc/classes/rpcserver.html
index 2bd9d0530..ab25fdef2 100644
--- a/docs/reference/api/typedoc/classes/rpcserver.html
+++ b/docs/reference/api/typedoc/classes/rpcserver.html
@@ -115,7 +115,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L92">rpc_server.ts:92</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L92">rpc_server.ts:92</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">get<wbr>Imports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">unknown</span><span class="tsd-signat [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
</ul>
</aside>
<div class="tsd-type-declaration">
@@ -201,7 +201,7 @@
<div class="tsd-signature tsd-kind-icon">key<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L78">rpc_server.ts:78</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L78">rpc_server.ts:78</a></li>
</ul>
</aside>
</section>
@@ -211,7 +211,7 @@
<div class="tsd-signature tsd-kind-icon">logger<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>msg<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
</ul>
</aside>
<div class="tsd-type-declaration">
@@ -242,7 +242,7 @@
<div class="tsd-signature tsd-kind-icon">socket<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">WebSocket</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
</ul>
</aside>
</section>
@@ -252,7 +252,7 @@
<div class="tsd-signature tsd-kind-icon">state<span class="tsd-signature-symbol">:</span> <a href="../enums/rpcserverstate.html" class="tsd-signature-type">RPCServerState</a><span class="tsd-signature-symbol"> = RPCServerState.InitHeader</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
</ul>
</aside>
</section>
@@ -262,7 +262,7 @@
<div class="tsd-signature tsd-kind-icon">url<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L77">rpc_server.ts:77</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L77">rpc_server.ts:77</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/classes/scalar.html b/docs/reference/api/typedoc/classes/scalar.html
index 651f48b03..7cce4f255 100644
--- a/docs/reference/api/typedoc/classes/scalar.html
+++ b/docs/reference/api/typedoc/classes/scalar.html
@@ -112,7 +112,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L145">runtime.ts:145</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -137,7 +137,7 @@
<div class="tsd-signature tsd-kind-icon">dtype<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L145">runtime.ts:145</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -152,7 +152,7 @@
<div class="tsd-signature tsd-kind-icon">value<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L143">runtime.ts:143</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L143">runtime.ts:143</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/webgpucontext.html b/docs/reference/api/typedoc/classes/webgpucontext.html
index 8653b2384..d65a38876 100644
--- a/docs/reference/api/typedoc/classes/webgpucontext.html
+++ b/docs/reference/api/typedoc/classes/webgpucontext.html
@@ -120,7 +120,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
</ul>
</aside>
<h4 class="tsd-parameters-title">Parameters</h4>
@@ -145,7 +145,7 @@
<div class="tsd-signature tsd-kind-icon">device<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">GPUDevice</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
</ul>
</aside>
</section>
@@ -155,7 +155,7 @@
<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
</ul>
</aside>
</section>
@@ -172,7 +172,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -209,7 +209,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/webgpu.ts#L170">webgpu.ts:170</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/webgpu.ts#L170">webgpu.ts:170</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -238,7 +238,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/enums/argtypecode.html b/docs/reference/api/typedoc/enums/argtypecode.html
index c65f2629c..72ea249b9 100644
--- a/docs/reference/api/typedoc/enums/argtypecode.html
+++ b/docs/reference/api/typedoc/enums/argtypecode.html
@@ -106,7 +106,7 @@
<div class="tsd-signature tsd-kind-icon">DLDevice<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 6</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L220">ctypes.ts:220</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L220">ctypes.ts:220</a></li>
</ul>
</aside>
</section>
@@ -116,7 +116,7 @@
<div class="tsd-signature tsd-kind-icon">Float<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L216">ctypes.ts:216</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L216">ctypes.ts:216</a></li>
</ul>
</aside>
</section>
@@ -126,7 +126,7 @@
<div class="tsd-signature tsd-kind-icon">Int<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 0</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L214">ctypes.ts:214</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L214">ctypes.ts:214</a></li>
</ul>
</aside>
</section>
@@ -136,7 +136,7 @@
<div class="tsd-signature tsd-kind-icon">Null<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L218">ctypes.ts:218</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L218">ctypes.ts:218</a></li>
</ul>
</aside>
</section>
@@ -146,7 +146,7 @@
<div class="tsd-signature tsd-kind-icon">TVMBytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 12</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L226">ctypes.ts:226</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L226">ctypes.ts:226</a></li>
</ul>
</aside>
</section>
@@ -156,7 +156,7 @@
<div class="tsd-signature tsd-kind-icon">TVMDLTensor<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 7</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L221">ctypes.ts:221</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L221">ctypes.ts:221</a></li>
</ul>
</aside>
</section>
@@ -166,7 +166,7 @@
<div class="tsd-signature tsd-kind-icon">TVMData<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 5</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L219">ctypes.ts:219</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L219">ctypes.ts:219</a></li>
</ul>
</aside>
</section>
@@ -176,7 +176,7 @@
<div class="tsd-signature tsd-kind-icon">TVMModule<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 9</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L223">ctypes.ts:223</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L223">ctypes.ts:223</a></li>
</ul>
</aside>
</section>
@@ -186,7 +186,7 @@
<div class="tsd-signature tsd-kind-icon">TVMNDArray<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 13</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L227">ctypes.ts:227</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L227">ctypes.ts:227</a></li>
</ul>
</aside>
</section>
@@ -196,7 +196,7 @@
<div class="tsd-signature tsd-kind-icon">TVMObject<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L222">ctypes.ts:222</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L222">ctypes.ts:222</a></li>
</ul>
</aside>
</section>
@@ -206,7 +206,7 @@
<div class="tsd-signature tsd-kind-icon">TVMObjectRValue<wbr>Ref<wbr>Arg<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 14</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L228">ctypes.ts:228</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L228">ctypes.ts:228</a></li>
</ul>
</aside>
</section>
@@ -216,7 +216,7 @@
<div class="tsd-signature tsd-kind-icon">TVMOpaque<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 3</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L217">ctypes.ts:217</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L217">ctypes.ts:217</a></li>
</ul>
</aside>
</section>
@@ -226,7 +226,7 @@
<div class="tsd-signature tsd-kind-icon">TVMPacked<wbr>Func<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 10</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L224">ctypes.ts:224</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L224">ctypes.ts:224</a></li>
</ul>
</aside>
</section>
@@ -236,7 +236,7 @@
<div class="tsd-signature tsd-kind-icon">TVMStr<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 11</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L225">ctypes.ts:225</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L225">ctypes.ts:225</a></li>
</ul>
</aside>
</section>
@@ -246,7 +246,7 @@
<div class="tsd-signature tsd-kind-icon">UInt<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/aynccallbackcode.html b/docs/reference/api/typedoc/enums/aynccallbackcode.html
index db44637ea..ae83dfbad 100644
--- a/docs/reference/api/typedoc/enums/aynccallbackcode.html
+++ b/docs/reference/api/typedoc/enums/aynccallbackcode.html
@@ -93,7 +93,7 @@
<div class="tsd-signature tsd-kind-icon">k<wbr>Exception<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 5</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L676">runtime.ts:676</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L676">runtime.ts:676</a></li>
</ul>
</aside>
</section>
@@ -103,7 +103,7 @@
<div class="tsd-signature tsd-kind-icon">k<wbr>Return<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L675">runtime.ts:675</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L675">runtime.ts:675</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/dldatatypecode.html b/docs/reference/api/typedoc/enums/dldatatypecode.html
index 236752108..d33b23d85 100644
--- a/docs/reference/api/typedoc/enums/dldatatypecode.html
+++ b/docs/reference/api/typedoc/enums/dldatatypecode.html
@@ -95,7 +95,7 @@
<div class="tsd-signature tsd-kind-icon">Float<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L242">runtime.ts:242</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L242">runtime.ts:242</a></li>
</ul>
</aside>
</section>
@@ -105,7 +105,7 @@
<div class="tsd-signature tsd-kind-icon">Int<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 0</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L240">runtime.ts:240</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L240">runtime.ts:240</a></li>
</ul>
</aside>
</section>
@@ -115,7 +115,7 @@
<div class="tsd-signature tsd-kind-icon">Opaque<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 3</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L243">runtime.ts:243</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L243">runtime.ts:243</a></li>
</ul>
</aside>
</section>
@@ -125,7 +125,7 @@
<div class="tsd-signature tsd-kind-icon">UInt<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L241">runtime.ts:241</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L241">runtime.ts:241</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/rpcserverstate.html b/docs/reference/api/typedoc/enums/rpcserverstate.html
index 306e633d6..9476bb9d7 100644
--- a/docs/reference/api/typedoc/enums/rpcserverstate.html
+++ b/docs/reference/api/typedoc/enums/rpcserverstate.html
@@ -90,7 +90,7 @@
<div class="tsd-signature tsd-kind-icon">Init<wbr>Header<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L27">rpc_server.ts:27</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L27">rpc_server.ts:27</a></li>
</ul>
</aside>
</section>
@@ -100,7 +100,7 @@
<div class="tsd-signature tsd-kind-icon">Init<wbr>Header<wbr>Key<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L28">rpc_server.ts:28</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L28">rpc_server.ts:28</a></li>
</ul>
</aside>
</section>
@@ -110,7 +110,7 @@
<div class="tsd-signature tsd-kind-icon">Init<wbr>Server<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L29">rpc_server.ts:29</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L29">rpc_server.ts:29</a></li>
</ul>
</aside>
</section>
@@ -120,7 +120,7 @@
<div class="tsd-signature tsd-kind-icon">Receive<wbr>Packet<wbr>Body<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L32">rpc_server.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L32">rpc_server.ts:32</a></li>
</ul>
</aside>
</section>
@@ -130,7 +130,7 @@
<div class="tsd-signature tsd-kind-icon">Receive<wbr>Packet<wbr>Header<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L31">rpc_server.ts:31</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L31">rpc_server.ts:31</a></li>
</ul>
</aside>
</section>
@@ -140,7 +140,7 @@
<div class="tsd-signature tsd-kind-icon">Wait<wbr>For<wbr>Callback<span class="tsd-signature-symbol">:</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/enums/sizeof.html b/docs/reference/api/typedoc/enums/sizeof.html
index f5ecf517f..5a8b02cf2 100644
--- a/docs/reference/api/typedoc/enums/sizeof.html
+++ b/docs/reference/api/typedoc/enums/sizeof.html
@@ -100,7 +100,7 @@
<div class="tsd-signature tsd-kind-icon">DLData<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = I32</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L206">ctypes.ts:206</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L206">ctypes.ts:206</a></li>
</ul>
</aside>
</section>
@@ -110,7 +110,7 @@
<div class="tsd-signature tsd-kind-icon">DLDevice<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = I32 + I32</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L207">ctypes.ts:207</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L207">ctypes.ts:207</a></li>
</ul>
</aside>
</section>
@@ -120,7 +120,7 @@
<div class="tsd-signature tsd-kind-icon">F32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L203">ctypes.ts:203</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L203">ctypes.ts:203</a></li>
</ul>
</aside>
</section>
@@ -130,7 +130,7 @@
<div class="tsd-signature tsd-kind-icon">F64<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L204">ctypes.ts:204</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L204">ctypes.ts:204</a></li>
</ul>
</aside>
</section>
@@ -140,7 +140,7 @@
<div class="tsd-signature tsd-kind-icon">I32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L201">ctypes.ts:201</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L201">ctypes.ts:201</a></li>
</ul>
</aside>
</section>
@@ -150,7 +150,7 @@
<div class="tsd-signature tsd-kind-icon">I64<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L202">ctypes.ts:202</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L202">ctypes.ts:202</a></li>
</ul>
</aside>
</section>
@@ -160,7 +160,7 @@
<div class="tsd-signature tsd-kind-icon">TVMValue<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L205">ctypes.ts:205</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L205">ctypes.ts:205</a></li>
</ul>
</aside>
</section>
@@ -170,7 +170,7 @@
<div class="tsd-signature tsd-kind-icon">U16<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L200">ctypes.ts:200</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L200">ctypes.ts:200</a></li>
</ul>
</aside>
</section>
@@ -180,7 +180,7 @@
<div class="tsd-signature tsd-kind-icon">U8<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L199">ctypes.ts:199</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L199">ctypes.ts:199</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/index.html b/docs/reference/api/typedoc/index.html
index 6b3b07119..f1feb70ad 100644
--- a/docs/reference/api/typedoc/index.html
+++ b/docs/reference/api/typedoc/index.html
@@ -174,7 +174,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Alloc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>shape<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, ndim<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, dtypeCode<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, dtypeBits<span class="tsd [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L112">ctypes.ts:112</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L112">ctypes.ts:112</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -238,7 +238,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>From<wbr>Bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, data<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nbytes<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">num [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L128">ctypes.ts:128</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L128">ctypes.ts:128</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -282,7 +282,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>From<wbr>To<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>from<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, to<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, stream<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-sig [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L144">ctypes.ts:144</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L144">ctypes.ts:144</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -326,7 +326,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>ToBytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, data<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nbytes<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</sp [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L136">ctypes.ts:136</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L136">ctypes.ts:136</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -370,7 +370,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L121">ctypes.ts:121</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L121">ctypes.ts:121</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -406,7 +406,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMBackend<wbr>PackedCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>argValues<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, argCodes<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nargs<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number< [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L160">ctypes.ts:160</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L160">ctypes.ts:160</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -458,7 +458,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMCFunc<wbr>Set<wbr>Return<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>ret<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, value<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCode<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signa [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L77">ctypes.ts:77</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L77">ctypes.ts:77</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -506,7 +506,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMCb<wbr>Arg<wbr>ToReturn<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>value<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, code<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span c [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L83">ctypes.ts:83</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L83">ctypes.ts:83</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -545,7 +545,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Call<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>func<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, argValues<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCode<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-t [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L67">ctypes.ts:67</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L67">ctypes.ts:67</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -601,7 +601,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>func<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L57">ctypes.ts:57</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L57">ctypes.ts:57</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -637,7 +637,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Get<wbr>Global<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>name<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span cla [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L100">ctypes.ts:100</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L100">ctypes.ts:100</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -676,7 +676,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>List<wbr>Global<wbr>Names<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>outSize<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, outArray<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&g [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L88">ctypes.ts:88</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L88">ctypes.ts:88</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -715,7 +715,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Register<wbr>Global<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>name<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, f<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, override<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</spa [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L94">ctypes.ts:94</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L94">ctypes.ts:94</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -758,7 +758,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMGet<wbr>Last<wbr>Error<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L34">ctypes.ts:34</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L34">ctypes.ts:34</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -788,7 +788,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L52">ctypes.ts:52</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L52">ctypes.ts:52</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -824,7 +824,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Get<wbr>Function<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, funcName<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, queryImports<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">numbe [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L42">ctypes.ts:42</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L42">ctypes.ts:42</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -872,7 +872,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Import<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, dep<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-si [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L48">ctypes.ts:48</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L48">ctypes.ts:48</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -912,7 +912,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMSynchronize<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>deviceType<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, deviceId<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, stream<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signatur [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L150">ctypes.ts:150</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L150">ctypes.ts:150</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -954,7 +954,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Alloc<wbr>Space<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>size<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L167">ctypes.ts:167</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L167">ctypes.ts:167</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -990,7 +990,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Free<wbr>Space<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>ptr<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L170">ctypes.ts:170</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L170">ctypes.ts:170</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1026,7 +1026,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Func<wbr>Create<wbr>FromCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>resource<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&g [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L187">ctypes.ts:187</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L187">ctypes.ts:187</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1066,7 +1066,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>PackedCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>args<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCodes<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nargs<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L179">ctypes.ts:179</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L179">ctypes.ts:179</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1118,7 +1118,7 @@
<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>PackedCFunc<wbr>Finalizer<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>resourceHandle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L193">ctypes.ts:193</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L193">ctypes.ts:193</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1154,7 +1154,7 @@
<div class="tsd-signature tsd-kind-icon">GPUPointer<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/webgpu.ts#L25">webgpu.ts:25</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/webgpu.ts#L25">webgpu.ts:25</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1169,7 +1169,7 @@
<div class="tsd-signature tsd-kind-icon">Packed<wbr>Func<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">...</span>args<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol"> & </span><a href="interfaces/disp [...]
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L36">runtime.ts:36</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L36">runtime.ts:36</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1184,7 +1184,7 @@
<div class="tsd-signature tsd-kind-icon">Pointer<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L25">ctypes.ts:25</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L25">ctypes.ts:25</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1199,7 +1199,7 @@
<div class="tsd-signature tsd-kind-icon">Ptr<wbr>Offset<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/ctypes.ts#L28">ctypes.ts:28</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/ctypes.ts#L28">ctypes.ts:28</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1217,7 +1217,7 @@
<div class="tsd-signature tsd-kind-icon">RPC_<wbr>MAGIC<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">1045105</span><span class="tsd-signature-symbol"> = 1045105</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/rpc_server.ts#L36">rpc_server.ts:36</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/rpc_server.ts#L36">rpc_server.ts:36</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1239,7 +1239,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/support.ts#L25">support.ts:25</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/support.ts#L25">support.ts:25</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1271,7 +1271,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/support.ts#L39">support.ts:39</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/support.ts#L39">support.ts:39</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1300,7 +1300,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/support.ts#L52">support.ts:52</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/support.ts#L52">support.ts:52</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1337,7 +1337,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/compact.ts#L38">compact.ts:38</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/compact.ts#L38">compact.ts:38</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1368,7 +1368,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/webgpu.ts#L30">webgpu.ts:30</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/webgpu.ts#L30">webgpu.ts:30</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1390,7 +1390,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/environment.ts#L32">environment.ts:32</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/environment.ts#L32">environment.ts:32</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1421,7 +1421,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/compact.ts#L24">compact.ts:24</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/compact.ts#L24">compact.ts:24</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1443,7 +1443,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L1356">runtime.ts:1356</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L1356">runtime.ts:1356</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1508,7 +1508,7 @@
<li class="tsd-description">
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/support.ts#L62">support.ts:62</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/support.ts#L62">support.ts:62</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -1530,7 +1530,7 @@
<div class="tsd-signature tsd-kind-icon">DLData<wbr>Type<wbr>Code<wbr>ToStr<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">object</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L246">runtime.ts:246</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L246">runtime.ts:246</a></li>
</ul>
</aside>
<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1539,7 +1539,7 @@
<div class="tsd-signature tsd-kind-icon">0<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "int"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L247">runtime.ts:247</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L247">runtime.ts:247</a></li>
</ul>
</aside>
</section>
@@ -1549,7 +1549,7 @@
<div class="tsd-signature tsd-kind-icon">1<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "uint"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L248">runtime.ts:248</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L248">runtime.ts:248</a></li>
</ul>
</aside>
</section>
@@ -1559,7 +1559,7 @@
<div class="tsd-signature tsd-kind-icon">2<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "float"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L249">runtime.ts:249</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L249">runtime.ts:249</a></li>
</ul>
</aside>
</section>
@@ -1569,7 +1569,7 @@
<div class="tsd-signature tsd-kind-icon">3<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "handle"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L250">runtime.ts:250</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L250">runtime.ts:250</a></li>
</ul>
</aside>
</section>
@@ -1580,7 +1580,7 @@
<div class="tsd-signature tsd-kind-icon">Device<wbr>Enum<wbr>ToStr<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">object</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L175">runtime.ts:175</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L175">runtime.ts:175</a></li>
</ul>
</aside>
<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1589,7 +1589,7 @@
<div class="tsd-signature tsd-kind-icon">1<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "cpu"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L176">runtime.ts:176</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L176">runtime.ts:176</a></li>
</ul>
</aside>
</section>
@@ -1599,7 +1599,7 @@
<div class="tsd-signature tsd-kind-icon">15<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "webgpu"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L180">runtime.ts:180</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L180">runtime.ts:180</a></li>
</ul>
</aside>
</section>
@@ -1609,7 +1609,7 @@
<div class="tsd-signature tsd-kind-icon">2<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "cuda"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L177">runtime.ts:177</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L177">runtime.ts:177</a></li>
</ul>
</aside>
</section>
@@ -1619,7 +1619,7 @@
<div class="tsd-signature tsd-kind-icon">4<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "opencl"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L178">runtime.ts:178</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L178">runtime.ts:178</a></li>
</ul>
</aside>
</section>
@@ -1629,7 +1629,7 @@
<div class="tsd-signature tsd-kind-icon">8<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = "metal"</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L179">runtime.ts:179</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L179">runtime.ts:179</a></li>
</ul>
</aside>
</section>
@@ -1640,7 +1640,7 @@
<div class="tsd-signature tsd-kind-icon">Device<wbr>Str<wbr>ToEnum<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">object</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L183">runtime.ts:183</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L183">runtime.ts:183</a></li>
</ul>
</aside>
<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1649,7 +1649,7 @@
<div class="tsd-signature tsd-kind-icon">cl<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L186">runtime.ts:186</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L186">runtime.ts:186</a></li>
</ul>
</aside>
</section>
@@ -1659,7 +1659,7 @@
<div class="tsd-signature tsd-kind-icon">cpu<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 1</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L184">runtime.ts:184</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L184">runtime.ts:184</a></li>
</ul>
</aside>
</section>
@@ -1669,7 +1669,7 @@
<div class="tsd-signature tsd-kind-icon">cuda<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 2</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L185">runtime.ts:185</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L185">runtime.ts:185</a></li>
</ul>
</aside>
</section>
@@ -1679,7 +1679,7 @@
<div class="tsd-signature tsd-kind-icon">metal<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 8</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L189">runtime.ts:189</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L189">runtime.ts:189</a></li>
</ul>
</aside>
</section>
@@ -1689,7 +1689,7 @@
<div class="tsd-signature tsd-kind-icon">opencl<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 4</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L187">runtime.ts:187</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L187">runtime.ts:187</a></li>
</ul>
</aside>
</section>
@@ -1699,7 +1699,7 @@
<div class="tsd-signature tsd-kind-icon">vulkan<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 7</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L188">runtime.ts:188</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L188">runtime.ts:188</a></li>
</ul>
</aside>
</section>
@@ -1709,7 +1709,7 @@
<div class="tsd-signature tsd-kind-icon">webgpu<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 15</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/runtime.ts#L190">runtime.ts:190</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/runtime.ts#L190">runtime.ts:190</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/interfaces/disposable.html b/docs/reference/api/typedoc/interfaces/disposable.html
index de6c82894..a1cef197b 100644
--- a/docs/reference/api/typedoc/interfaces/disposable.html
+++ b/docs/reference/api/typedoc/interfaces/disposable.html
@@ -113,7 +113,7 @@
<div class="tsd-signature tsd-kind-icon">dispose<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/types.ts#L52">types.ts:52</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/types.ts#L52">types.ts:52</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/interfaces/functioninfo.html b/docs/reference/api/typedoc/interfaces/functioninfo.html
index 4cdf70b7f..106e46f1a 100644
--- a/docs/reference/api/typedoc/interfaces/functioninfo.html
+++ b/docs/reference/api/typedoc/interfaces/functioninfo.html
@@ -95,7 +95,7 @@
<div class="tsd-signature tsd-kind-icon">arg_<wbr>types<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/webgpu.ts#L41">webgpu.ts:41</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/webgpu.ts#L41">webgpu.ts:41</a></li>
</ul>
</aside>
</section>
@@ -105,7 +105,7 @@
<div class="tsd-signature tsd-kind-icon">launch_<wbr>param_<wbr>tags<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/webgpu.ts#L42">webgpu.ts:42</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/webgpu.ts#L42">webgpu.ts:42</a></li>
</ul>
</aside>
</section>
@@ -115,7 +115,7 @@
<div class="tsd-signature tsd-kind-icon">name<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/webgpu.ts#L40">webgpu.ts:40</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/webgpu.ts#L40">webgpu.ts:40</a></li>
</ul>
</aside>
</section>
diff --git a/docs/reference/api/typedoc/interfaces/libraryprovider.html b/docs/reference/api/typedoc/interfaces/libraryprovider.html
index e6b04474f..b589fd867 100644
--- a/docs/reference/api/typedoc/interfaces/libraryprovider.html
+++ b/docs/reference/api/typedoc/interfaces/libraryprovider.html
@@ -112,7 +112,7 @@
<div class="tsd-signature tsd-kind-icon">imports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol"><</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol">></span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/types.ts#L34">types.ts:34</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/types.ts#L34">types.ts:34</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
@@ -127,7 +127,7 @@
<div class="tsd-signature tsd-kind-icon">start<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>inst<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">Instance</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> => </span><span class="tsd-signature-type">void</span></div>
<aside class="tsd-sources">
<ul>
- <li>Defined in <a href="https://github.com/apache/tvm/blob/8691cbed0/web/src/types.ts#L39">types.ts:39</a></li>
+ <li>Defined in <a href="https://github.com/apache/tvm/blob/bce57586b/web/src/types.ts#L39">types.ts:39</a></li>
</ul>
</aside>
<div class="tsd-comment tsd-typography">
diff --git a/docs/searchindex.js b/docs/searchindex.js
index d42883545..8999078d4 100644
--- a/docs/searchindex.js
+++ b/docs/searchindex.js
@@ -1 +1 @@
-Search.setIndex({docnames:["arch/benchmark","arch/convert_layout","arch/debugger","arch/device_target_interactions","arch/frontend/tensorflow","arch/hybrid_script","arch/index","arch/inferbound","arch/introduction_to_module_serialization","arch/microtvm_design","arch/microtvm_project_api","arch/model_library_format","arch/pass_infra","arch/relay_intro","arch/relay_op_strategy","arch/runtime","arch/runtimes/vulkan","arch/security","arch/virtual_machine","contribute/ci","contribute/code_gu [...]
\ No newline at end of file
+Search.setIndex({docnames:["arch/benchmark","arch/convert_layout","arch/debugger","arch/device_target_interactions","arch/frontend/tensorflow","arch/hybrid_script","arch/index","arch/inferbound","arch/introduction_to_module_serialization","arch/microtvm_design","arch/microtvm_project_api","arch/model_library_format","arch/pass_infra","arch/relay_intro","arch/relay_op_strategy","arch/runtime","arch/runtimes/vulkan","arch/security","arch/virtual_machine","contribute/ci","contribute/code_gu [...]
\ No newline at end of file
diff --git a/docs/topic/vta/tutorials/autotvm/sg_execution_times.html b/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
index 1c1c52b35..78fb28ae4 100644
--- a/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
@@ -300,10 +300,10 @@
<div class="section" id="computation-times">
<span id="sphx-glr-topic-vta-tutorials-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:20.608</strong> total execution time for <strong>topic_vta_tutorials_autotvm</strong> files:</p>
+<p><strong>00:20.754</strong> total execution time for <strong>topic_vta_tutorials_autotvm</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:20.405</strong>: <a class="reference internal" href="tune_relay_vta.html#sphx-glr-topic-vta-tutorials-autotvm-tune-relay-vta-py"><span class="std std-ref">Auto-tuning a convolutional network on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_vta.py</span></code>)</p></li>
-<li><p><strong>00:00.203</strong>: <a class="reference internal" href="tune_alu_vta.html#sphx-glr-topic-vta-tutorials-autotvm-tune-alu-vta-py"><span class="std std-ref">Auto-tuning a ALU fused op on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_alu_vta.py</span></code>)</p></li>
+<li><p><strong>00:20.557</strong>: <a class="reference internal" href="tune_relay_vta.html#sphx-glr-topic-vta-tutorials-autotvm-tune-relay-vta-py"><span class="std std-ref">Auto-tuning a convolutional network on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_vta.py</span></code>)</p></li>
+<li><p><strong>00:00.197</strong>: <a class="reference internal" href="tune_alu_vta.html#sphx-glr-topic-vta-tutorials-autotvm-tune-alu-vta-py"><span class="std std-ref">Auto-tuning a ALU fused op on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_alu_vta.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/topic/vta/tutorials/frontend/deploy_classification.html b/docs/topic/vta/tutorials/frontend/deploy_classification.html
index c88fbdb62..daf62afe3 100644
--- a/docs/topic/vta/tutorials/frontend/deploy_classification.html
+++ b/docs/topic/vta/tutorials/frontend/deploy_classification.html
@@ -539,7 +539,7 @@ and dense layer which will both be executed in fp32 on the CPU.</p></li>
DeprecationWarning,
/workspace/vta/tutorials/frontend/deploy_classification.py:213: DeprecationWarning: legacy graph executor behavior of producing json / lib / params will be removed in the next release. Please see documents of tvm.contrib.graph_executor.GraphModule for the new recommended usage.
relay_prog, target=tvm.target.Target(target, host=env.target_host), params=params
-resnet18_v1 inference graph built in 21.57s!
+resnet18_v1 inference graph built in 20.94s!
</pre></div>
</div>
</div>
diff --git a/docs/topic/vta/tutorials/frontend/deploy_detection.html b/docs/topic/vta/tutorials/frontend/deploy_detection.html
index 30bf049c4..980dd92d0 100644
--- a/docs/topic/vta/tutorials/frontend/deploy_detection.html
+++ b/docs/topic/vta/tutorials/frontend/deploy_detection.html
@@ -557,7 +557,7 @@ and dense layer which will both be executed in fp32 on the CPU.</p></li>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/workspace/python/tvm/relay/build_module.py:439: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
DeprecationWarning,
-yolov3-tiny inference graph built in 14.97s!
+yolov3-tiny inference graph built in 14.56s!
</pre></div>
</div>
</div>
diff --git a/docs/topic/vta/tutorials/frontend/sg_execution_times.html b/docs/topic/vta/tutorials/frontend/sg_execution_times.html
index 0c1bf6026..c036ca4da 100644
--- a/docs/topic/vta/tutorials/frontend/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/frontend/sg_execution_times.html
@@ -300,10 +300,10 @@
<div class="section" id="computation-times">
<span id="sphx-glr-topic-vta-tutorials-frontend-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>01:27.486</strong> total execution time for <strong>topic_vta_tutorials_frontend</strong> files:</p>
+<p><strong>01:27.989</strong> total execution time for <strong>topic_vta_tutorials_frontend</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:46.206</strong>: <a class="reference internal" href="deploy_detection.html#sphx-glr-topic-vta-tutorials-frontend-deploy-detection-py"><span class="std std-ref">Deploy Pretrained Vision Detection Model from Darknet on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_detection.py</span></code>)</p></li>
-<li><p><strong>00:41.280</strong>: <a class="reference internal" href="deploy_classification.html#sphx-glr-topic-vta-tutorials-frontend-deploy-classification-py"><span class="std std-ref">Deploy Pretrained Vision Model from MxNet on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_classification.py</span></code>)</p></li>
+<li><p><strong>00:46.818</strong>: <a class="reference internal" href="deploy_detection.html#sphx-glr-topic-vta-tutorials-frontend-deploy-detection-py"><span class="std std-ref">Deploy Pretrained Vision Detection Model from Darknet on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_detection.py</span></code>)</p></li>
+<li><p><strong>00:41.170</strong>: <a class="reference internal" href="deploy_classification.html#sphx-glr-topic-vta-tutorials-frontend-deploy-classification-py"><span class="std std-ref">Deploy Pretrained Vision Model from MxNet on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_classification.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/topic/vta/tutorials/optimize/sg_execution_times.html b/docs/topic/vta/tutorials/optimize/sg_execution_times.html
index a70bdd6f5..a830bb4bf 100644
--- a/docs/topic/vta/tutorials/optimize/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/optimize/sg_execution_times.html
@@ -300,10 +300,10 @@
<div class="section" id="computation-times">
<span id="sphx-glr-topic-vta-tutorials-optimize-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:03.472</strong> total execution time for <strong>topic_vta_tutorials_optimize</strong> files:</p>
+<p><strong>00:03.767</strong> total execution time for <strong>topic_vta_tutorials_optimize</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:02.936</strong>: <a class="reference internal" href="convolution_opt.html#sphx-glr-topic-vta-tutorials-optimize-convolution-opt-py"><span class="std std-ref">2D Convolution Optimization</span></a> (<code class="docutils literal notranslate"><span class="pre">convolution_opt.py</span></code>)</p></li>
-<li><p><strong>00:00.537</strong>: <a class="reference internal" href="matrix_multiply_opt.html#sphx-glr-topic-vta-tutorials-optimize-matrix-multiply-opt-py"><span class="std std-ref">Matrix Multiply Blocking</span></a> (<code class="docutils literal notranslate"><span class="pre">matrix_multiply_opt.py</span></code>)</p></li>
+<li><p><strong>00:03.229</strong>: <a class="reference internal" href="convolution_opt.html#sphx-glr-topic-vta-tutorials-optimize-convolution-opt-py"><span class="std std-ref">2D Convolution Optimization</span></a> (<code class="docutils literal notranslate"><span class="pre">convolution_opt.py</span></code>)</p></li>
+<li><p><strong>00:00.538</strong>: <a class="reference internal" href="matrix_multiply_opt.html#sphx-glr-topic-vta-tutorials-optimize-matrix-multiply-opt-py"><span class="std std-ref">Matrix Multiply Blocking</span></a> (<code class="docutils literal notranslate"><span class="pre">matrix_multiply_opt.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/topic/vta/tutorials/sg_execution_times.html b/docs/topic/vta/tutorials/sg_execution_times.html
index 593998d14..539a089d3 100644
--- a/docs/topic/vta/tutorials/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/sg_execution_times.html
@@ -300,10 +300,10 @@
<div class="section" id="computation-times">
<span id="sphx-glr-topic-vta-tutorials-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:00.971</strong> total execution time for <strong>topic_vta_tutorials</strong> files:</p>
+<p><strong>00:00.976</strong> total execution time for <strong>topic_vta_tutorials</strong> files:</p>
<ul class="simple">
-<li><p><strong>00:00.494</strong>: <a class="reference internal" href="matrix_multiply.html#sphx-glr-topic-vta-tutorials-matrix-multiply-py"><span class="std std-ref">Simple Matrix Multiply</span></a> (<code class="docutils literal notranslate"><span class="pre">matrix_multiply.py</span></code>)</p></li>
-<li><p><strong>00:00.476</strong>: <a class="reference internal" href="vta_get_started.html#sphx-glr-topic-vta-tutorials-vta-get-started-py"><span class="std std-ref">Get Started with VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">vta_get_started.py</span></code>)</p></li>
+<li><p><strong>00:00.499</strong>: <a class="reference internal" href="matrix_multiply.html#sphx-glr-topic-vta-tutorials-matrix-multiply-py"><span class="std std-ref">Simple Matrix Multiply</span></a> (<code class="docutils literal notranslate"><span class="pre">matrix_multiply.py</span></code>)</p></li>
+<li><p><strong>00:00.477</strong>: <a class="reference internal" href="vta_get_started.html#sphx-glr-topic-vta-tutorials-vta-get-started-py"><span class="std std-ref">Get Started with VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">vta_get_started.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/tutorial/auto_scheduler_matmul_x86.html b/docs/tutorial/auto_scheduler_matmul_x86.html
index f7603e0b1..b60040e79 100644
--- a/docs/tutorial/auto_scheduler_matmul_x86.html
+++ b/docs/tutorial/auto_scheduler_matmul_x86.html
@@ -545,7 +545,7 @@ operator fusion.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 93.500 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 93.729 ms
</pre></div>
</div>
</div>
@@ -611,7 +611,6 @@ resume the status and do more 5 trials.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Resume search:
/usr/local/lib/python3.7/dist-packages/xgboost/training.py:17: UserWarning: Old style callback is deprecated. See: https://xgboost.readthedocs.io/en/latest/python/callbacks.html
warnings.warn(f'Old style callback is deprecated. See: {link}', UserWarning)
-*E
</pre></div>
</div>
</div>
@@ -622,7 +621,7 @@ automatically optimize a matrix multiplication, without the need to specify a
search template. It ends a series of examples that starts from the Tensor
Expression (TE) language that demonstrates how TVM can optimize computational
operations.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 7.232 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 3.810 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-tutorial-auto-scheduler-matmul-x86-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../_downloads/eac4389b114db015e95cb3cdf8b86b83/auto_scheduler_matmul_x86.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">auto_scheduler_matmul_x86.py</span></code></a></p>
diff --git a/docs/tutorial/autotvm_relay_x86.html b/docs/tutorial/autotvm_relay_x86.html
index fa7228fd3..cf486b6cf 100644
--- a/docs/tutorial/autotvm_relay_x86.html
+++ b/docs/tutorial/autotvm_relay_x86.html
@@ -513,7 +513,7 @@ standard deviation.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{'mean': 496.60638313000044, 'median': 496.4213516499967, 'std': 0.8980274498640965}
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{'mean': 492.6316053099993, 'median': 492.4662148999971, 'std': 0.45000288160307683}
</pre></div>
</div>
</div>
@@ -667,129 +667,129 @@ depending on the specifics of the model and the target platform.</p>
</div>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[Task 1/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 1/25] Current/Best: 9.89/ 12.02 GFLOPS | Progress: (4/10) | 6.34 s
-[Task 1/25] Current/Best: 8.53/ 23.62 GFLOPS | Progress: (8/10) | 8.48 s
-[Task 1/25] Current/Best: 10.63/ 23.62 GFLOPS | Progress: (10/10) | 9.56 s Done.
+[Task 1/25] Current/Best: 17.15/ 23.91 GFLOPS | Progress: (4/10) | 4.81 s
+[Task 1/25] Current/Best: 5.81/ 23.91 GFLOPS | Progress: (8/10) | 8.61 s
+[Task 1/25] Current/Best: 16.74/ 23.91 GFLOPS | Progress: (10/10) | 9.73 s Done.
[Task 2/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 2/25] Current/Best: 7.44/ 20.07 GFLOPS | Progress: (4/10) | 2.11 s
-[Task 2/25] Current/Best: 15.55/ 20.07 GFLOPS | Progress: (8/10) | 3.19 s
-[Task 2/25] Current/Best: 22.68/ 22.68 GFLOPS | Progress: (10/10) | 3.62 s Done.
+[Task 2/25] Current/Best: 17.55/ 17.55 GFLOPS | Progress: (4/10) | 2.70 s
+[Task 2/25] Current/Best: 14.91/ 20.17 GFLOPS | Progress: (8/10) | 3.87 s
+[Task 2/25] Current/Best: 20.68/ 20.68 GFLOPS | Progress: (10/10) | 4.35 s Done.
[Task 3/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 3/25] Current/Best: 24.13/ 24.13 GFLOPS | Progress: (4/10) | 2.99 s
-[Task 3/25] Current/Best: 12.03/ 24.13 GFLOPS | Progress: (8/10) | 7.25 s
-[Task 3/25] Current/Best: 16.98/ 24.13 GFLOPS | Progress: (10/10) | 8.16 s Done.
+[Task 3/25] Current/Best: 17.86/ 18.99 GFLOPS | Progress: (4/10) | 2.68 s
+[Task 3/25] Current/Best: 7.63/ 22.54 GFLOPS | Progress: (8/10) | 4.62 s
+[Task 3/25] Current/Best: 14.78/ 22.54 GFLOPS | Progress: (10/10) | 5.56 s Done.
[Task 4/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 4/25] Current/Best: 14.55/ 17.66 GFLOPS | Progress: (4/10) | 3.01 s
-[Task 4/25] Current/Best: 10.05/ 17.66 GFLOPS | Progress: (8/10) | 4.51 s
-[Task 4/25] Current/Best: 16.66/ 17.66 GFLOPS | Progress: (10/10) | 5.20 s Done.
+[Task 4/25] Current/Best: 11.82/ 14.62 GFLOPS | Progress: (4/10) | 5.69 s
+[Task 4/25] Current/Best: 13.01/ 14.71 GFLOPS | Progress: (8/10) | 7.52 s
+[Task 4/25] Current/Best: 14.33/ 16.94 GFLOPS | Progress: (10/10) | 8.30 s Done.
[Task 5/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 5/25] Current/Best: 1.71/ 18.05 GFLOPS | Progress: (4/10) | 4.45 s
-[Task 5/25] Current/Best: 9.74/ 18.05 GFLOPS | Progress: (8/10) | 6.92 s
-[Task 5/25] Current/Best: 18.35/ 18.35 GFLOPS | Progress: (10/10) | 7.64 s Done.
+[Task 5/25] Current/Best: 11.36/ 16.91 GFLOPS | Progress: (4/10) | 2.63 s
+[Task 5/25] Current/Best: 5.16/ 19.29 GFLOPS | Progress: (8/10) | 4.47 s
+[Task 5/25] Current/Best: 9.76/ 19.29 GFLOPS | Progress: (10/10) | 5.14 s Done.
[Task 6/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 6/25] Current/Best: 9.62/ 21.76 GFLOPS | Progress: (4/10) | 3.03 s
-[Task 6/25] Current/Best: 16.67/ 21.76 GFLOPS | Progress: (8/10) | 4.75 s
-[Task 6/25] Current/Best: 13.17/ 21.76 GFLOPS | Progress: (10/10) | 6.49 s Done.
+[Task 6/25] Current/Best: 16.33/ 16.33 GFLOPS | Progress: (4/10) | 3.11 s
+[Task 6/25] Current/Best: 10.78/ 16.33 GFLOPS | Progress: (8/10) | 6.33 s
+[Task 6/25] Current/Best: 13.49/ 18.21 GFLOPS | Progress: (10/10) | 7.05 s Done.
[Task 7/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 7/25] Current/Best: 12.34/ 15.54 GFLOPS | Progress: (4/10) | 3.80 s
-[Task 7/25] Current/Best: 9.77/ 17.94 GFLOPS | Progress: (8/10) | 6.40 s
-[Task 7/25] Current/Best: 1.58/ 17.94 GFLOPS | Progress: (10/10) | 8.66 s Done.
+[Task 7/25] Current/Best: 10.59/ 21.24 GFLOPS | Progress: (4/10) | 2.67 s
+[Task 7/25] Current/Best: 6.48/ 21.24 GFLOPS | Progress: (8/10) | 4.80 s
+[Task 7/25] Current/Best: 6.72/ 21.24 GFLOPS | Progress: (10/10) | 5.96 s Done.
[Task 8/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 8/25] Current/Best: 19.10/ 19.10 GFLOPS | Progress: (4/10) | 3.08 s
-[Task 8/25] Current/Best: 10.45/ 19.10 GFLOPS | Progress: (8/10) | 5.11 s
-[Task 8/25] Current/Best: 10.46/ 19.10 GFLOPS | Progress: (10/10) | 6.74 s Done.
+[Task 8/25] Current/Best: 12.71/ 22.83 GFLOPS | Progress: (4/10) | 2.95 s
+[Task 8/25] Current/Best: 8.97/ 22.83 GFLOPS | Progress: (8/10) | 5.86 s
+[Task 8/25] Current/Best: 5.73/ 22.83 GFLOPS | Progress: (10/10) | 6.93 s Done.
[Task 9/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 9/25] Current/Best: 17.12/ 22.42 GFLOPS | Progress: (4/10) | 2.43 s
-[Task 9/25] Current/Best: 15.09/ 22.42 GFLOPS | Progress: (8/10) | 3.85 s
-[Task 9/25] Current/Best: 20.70/ 22.42 GFLOPS | Progress: (10/10) | 6.31 s Done.
+[Task 9/25] Current/Best: 9.99/ 20.35 GFLOPS | Progress: (4/10) | 4.60 s
+[Task 9/25] Current/Best: 11.69/ 20.35 GFLOPS | Progress: (8/10) | 6.14 s
+[Task 9/25] Current/Best: 12.63/ 20.35 GFLOPS | Progress: (10/10) | 15.62 s Done.
[Task 10/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 10/25] Current/Best: 13.05/ 13.05 GFLOPS | Progress: (4/10) | 2.34 s
-[Task 10/25] Current/Best: 9.54/ 20.94 GFLOPS | Progress: (8/10) | 3.91 s
-[Task 10/25] Current/Best: 16.56/ 20.94 GFLOPS | Progress: (10/10) | 4.81 s Done.
+[Task 10/25] Current/Best: 14.13/ 20.33 GFLOPS | Progress: (4/10) | 2.35 s
+[Task 10/25] Current/Best: 19.04/ 20.33 GFLOPS | Progress: (8/10) | 4.01 s
+[Task 10/25] Current/Best: 19.21/ 20.33 GFLOPS | Progress: (10/10) | 5.13 s Done.
[Task 11/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 11/25] Current/Best: 14.02/ 21.31 GFLOPS | Progress: (4/10) | 3.29 s
-[Task 11/25] Current/Best: 18.81/ 23.74 GFLOPS | Progress: (8/10) | 5.37 s
-[Task 11/25] Current/Best: 19.17/ 23.74 GFLOPS | Progress: (10/10) | 6.23 s Done.
+[Task 11/25] Current/Best: 18.01/ 18.01 GFLOPS | Progress: (4/10) | 3.35 s
+[Task 11/25] Current/Best: 12.37/ 20.88 GFLOPS | Progress: (8/10) | 5.39 s
+[Task 11/25] Current/Best: 1.59/ 20.88 GFLOPS | Progress: (10/10) | 7.66 s Done.
[Task 12/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 12/25] Current/Best: 16.29/ 16.29 GFLOPS | Progress: (4/10) | 3.15 s
-[Task 12/25] Current/Best: 14.14/ 22.82 GFLOPS | Progress: (8/10) | 4.75 s
-[Task 12/25] Current/Best: 8.12/ 22.82 GFLOPS | Progress: (10/10) | 6.28 s Done.
+[Task 12/25] Current/Best: 10.73/ 20.18 GFLOPS | Progress: (4/10) | 2.98 s
+[Task 12/25] Current/Best: 13.21/ 20.18 GFLOPS | Progress: (8/10) | 6.33 s
+[Task 12/25] Current/Best: 16.36/ 20.18 GFLOPS | Progress: (10/10) | 7.35 s Done.
[Task 13/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 13/25] Current/Best: 7.80/ 13.32 GFLOPS | Progress: (4/10) | 4.27 s
-[Task 13/25] Current/Best: 23.80/ 23.89 GFLOPS | Progress: (8/10) | 6.90 s
-[Task 13/25] Current/Best: 16.79/ 23.89 GFLOPS | Progress: (10/10) | 7.80 s Done.
+[Task 13/25] Current/Best: 11.47/ 11.47 GFLOPS | Progress: (4/10) | 5.08 s
+[Task 13/25] Current/Best: 4.74/ 15.17 GFLOPS | Progress: (8/10) | 8.45 s
+[Task 13/25] Current/Best: 18.73/ 18.73 GFLOPS | Progress: (10/10) | 10.62 s Done.
[Task 14/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 14/25] Current/Best: 13.94/ 15.06 GFLOPS | Progress: (4/10) | 3.43 s
-[Task 14/25] Current/Best: 16.27/ 16.27 GFLOPS | Progress: (8/10) | 4.95 s
-[Task 14/25] Current/Best: 10.95/ 16.27 GFLOPS | Progress: (10/10) | 6.05 s
-[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 15/25] Current/Best: 13.55/ 13.55 GFLOPS | Progress: (4/10) | 3.07 s
-[Task 15/25] Current/Best: 3.17/ 13.55 GFLOPS | Progress: (8/10) | 6.84 s
-[Task 15/25] Current/Best: 20.67/ 20.67 GFLOPS | Progress: (10/10) | 9.12 s
-[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s Done.
- Done.
+[Task 14/25] Current/Best: 14.51/ 16.34 GFLOPS | Progress: (4/10) | 3.94 s
+[Task 14/25] Current/Best: 7.15/ 16.34 GFLOPS | Progress: (8/10) | 7.10 s
+[Task 14/25] Current/Best: 13.16/ 19.19 GFLOPS | Progress: (10/10) | 7.86 s Done.
-[Task 16/25] Current/Best: 18.23/ 18.23 GFLOPS | Progress: (4/10) | 2.45 s
-[Task 16/25] Current/Best: 10.45/ 18.23 GFLOPS | Progress: (8/10) | 4.22 s
-[Task 16/25] Current/Best: 18.05/ 18.23 GFLOPS | Progress: (10/10) | 4.96 s Done.
+[Task 15/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
+[Task 15/25] Current/Best: 10.18/ 23.21 GFLOPS | Progress: (4/10) | 2.91 s
+[Task 15/25] Current/Best: 17.88/ 23.21 GFLOPS | Progress: (8/10) | 4.53 s
+[Task 15/25] Current/Best: 18.97/ 23.21 GFLOPS | Progress: (10/10) | 5.31 s
+[Task 16/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
+[Task 16/25] Current/Best: 15.88/ 15.88 GFLOPS | Progress: (4/10) | 2.91 s
+[Task 16/25] Current/Best: 6.17/ 15.88 GFLOPS | Progress: (8/10) | 4.91 s
+[Task 16/25] Current/Best: 10.18/ 15.88 GFLOPS | Progress: (10/10) | 7.78 s Done.
[Task 17/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 17/25] Current/Best: 6.22/ 22.92 GFLOPS | Progress: (4/10) | 2.75 s
-[Task 17/25] Current/Best: 6.15/ 22.92 GFLOPS | Progress: (8/10) | 5.02 s
-[Task 17/25] Current/Best: 10.99/ 22.92 GFLOPS | Progress: (10/10) | 6.45 s Done.
+[Task 17/25] Current/Best: 20.55/ 20.55 GFLOPS | Progress: (4/10) | 3.11 s
+[Task 17/25] Current/Best: 10.25/ 20.55 GFLOPS | Progress: (8/10) | 6.04 s
+[Task 17/25] Current/Best: 7.50/ 20.55 GFLOPS | Progress: (10/10) | 8.26 s Done.
[Task 18/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 18/25] Current/Best: 7.75/ 20.47 GFLOPS | Progress: (4/10) | 3.10 s
-[Task 18/25] Current/Best: 19.60/ 20.47 GFLOPS | Progress: (8/10) | 4.72 s
-[Task 18/25] Current/Best: 14.31/ 20.47 GFLOPS | Progress: (10/10) | 9.32 s Done.
+[Task 18/25] Current/Best: 9.55/ 19.41 GFLOPS | Progress: (4/10) | 3.16 s
+[Task 18/25] Current/Best: 17.21/ 22.12 GFLOPS | Progress: (8/10) | 4.74 s
+[Task 18/25] Current/Best: 16.47/ 22.12 GFLOPS | Progress: (10/10) | 5.95 s Done.
[Task 19/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 19/25] Current/Best: 9.19/ 19.04 GFLOPS | Progress: (4/10) | 5.30 s
-[Task 19/25] Current/Best: 18.99/ 19.04 GFLOPS | Progress: (8/10) | 7.15 s
-[Task 19/25] Current/Best: 18.66/ 19.04 GFLOPS | Progress: (10/10) | 8.54 s Done.
+[Task 19/25] Current/Best: 10.75/ 21.84 GFLOPS | Progress: (4/10) | 3.63 s
+[Task 19/25] Current/Best: 5.30/ 21.84 GFLOPS | Progress: (8/10) | 7.32 s
+[Task 19/25] Current/Best: 22.77/ 22.77 GFLOPS | Progress: (10/10) | 8.56 s Done.
[Task 20/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 20/25] Current/Best: 3.12/ 17.86 GFLOPS | Progress: (4/10) | 2.73 s
-[Task 20/25] Current/Best: 20.73/ 20.73 GFLOPS | Progress: (8/10) | 4.70 s
-[Task 20/25] Current/Best: 17.47/ 20.73 GFLOPS | Progress: (10/10) | 6.85 s
-[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 21/25] Current/Best: 10.86/ 14.28 GFLOPS | Progress: (4/10) | 2.95 s
-[Task 21/25] Current/Best: 13.95/ 20.94 GFLOPS | Progress: (8/10) | 7.91 s
-[Task 21/25] Current/Best: 16.36/ 20.94 GFLOPS | Progress: (10/10) | 8.43 s
+[Task 20/25] Current/Best: 5.34/ 16.10 GFLOPS | Progress: (4/10) | 3.22 s
+[Task 20/25] Current/Best: 8.96/ 16.10 GFLOPS | Progress: (8/10) | 4.64 s
+[Task 20/25] Current/Best: 5.19/ 16.10 GFLOPS | Progress: (10/10) | 9.68 s
+[Task 21/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s Done.
+ Done.
+
+[Task 21/25] Current/Best: 7.00/ 17.83 GFLOPS | Progress: (4/10) | 3.10 s
+[Task 21/25] Current/Best: 8.89/ 23.44 GFLOPS | Progress: (8/10) | 4.91 s
+[Task 21/25] Current/Best: 0.00/ 23.44 GFLOPS | Progress: (10/10) | 5.53 s
[Task 22/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 22/25] Current/Best: 10.48/ 14.86 GFLOPS | Progress: (4/10) | 3.42 s
-[Task 22/25] Current/Best: 15.16/ 16.42 GFLOPS | Progress: (8/10) | 6.03 s
-[Task 22/25] Current/Best: 8.44/ 16.42 GFLOPS | Progress: (10/10) | 7.37 s Done.
+[Task 22/25] Current/Best: 18.58/ 21.71 GFLOPS | Progress: (4/10) | 2.65 s
+[Task 22/25] Current/Best: 14.79/ 21.71 GFLOPS | Progress: (8/10) | 4.44 s
+[Task 22/25] Current/Best: 5.94/ 21.71 GFLOPS | Progress: (10/10) | 6.47 s Done.
[Task 23/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 23/25] Current/Best: 12.10/ 14.00 GFLOPS | Progress: (4/10) | 3.77 s
-[Task 23/25] Current/Best: 19.71/ 19.71 GFLOPS | Progress: (8/10) | 7.02 s
-[Task 23/25] Current/Best: 10.81/ 19.71 GFLOPS | Progress: (10/10) | 7.99 s Done.
+[Task 23/25] Current/Best: 22.80/ 22.80 GFLOPS | Progress: (4/10) | 5.67 s
+[Task 23/25] Current/Best: 20.16/ 22.80 GFLOPS | Progress: (8/10) | 8.49 s
+[Task 23/25] Current/Best: 7.01/ 22.80 GFLOPS | Progress: (10/10) | 10.75 s Done.
[Task 24/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s
-[Task 24/25] Current/Best: 4.07/ 4.07 GFLOPS | Progress: (4/10) | 5.50 s
-[Task 24/25] Current/Best: 3.38/ 8.30 GFLOPS | Progress: (8/10) | 18.01 s
-[Task 24/25] Current/Best: 0.00/ 8.30 GFLOPS | Progress: (10/10) | 228.79 s
+[Task 24/25] Current/Best: 1.97/ 4.99 GFLOPS | Progress: (4/10) | 12.00 s
+[Task 24/25] Current/Best: 8.84/ 10.59 GFLOPS | Progress: (8/10) | 24.16 s
+[Task 24/25] Current/Best: 1.15/ 10.59 GFLOPS | Progress: (10/10) | 26.93 s
[Task 25/25] Current/Best: 0.00/ 0.00 GFLOPS | Progress: (0/10) | 0.00 s Done.
Done.
- Done.
-[Task 25/25] Current/Best: 6.37/ 7.19 GFLOPS | Progress: (4/10) | 4.10 s
-[Task 25/25] Current/Best: 1.55/ 9.03 GFLOPS | Progress: (8/10) | 6.59 s
-[Task 25/25] Current/Best: 10.12/ 10.12 GFLOPS | Progress: (10/10) | 19.14 s
+[Task 25/25] Current/Best: 5.10/ 8.37 GFLOPS | Progress: (4/10) | 3.96 s
+[Task 25/25] Current/Best: 3.50/ 8.37 GFLOPS | Progress: (8/10) | 12.39 s
+[Task 25/25] Current/Best: 1.48/ 9.49 GFLOPS | Progress: (10/10) | 17.37 s Done.
</pre></div>
</div>
<p>The output from this tuning process will look something like this:</p>
@@ -890,8 +890,8 @@ improvement in comparing the optimized model to the unoptimized model.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>optimized: {'mean': 425.2909051199981, 'median': 425.08289284999137, 'std': 0.6725116699840613}
-unoptimized: {'mean': 496.60638313000044, 'median': 496.4213516499967, 'std': 0.8980274498640965}
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>optimized: {'mean': 433.3818005000001, 'median': 433.01172650000126, 'std': 1.2187799667414965}
+unoptimized: {'mean': 492.6316053099993, 'median': 492.4662148999971, 'std': 0.45000288160307683}
</pre></div>
</div>
</div>
@@ -905,7 +905,7 @@ models.</p>
<p>Here we presented a simple example using ResNet-50 v2 locally. However, TVM
supports many more features including cross-compilation, remote execution and
profiling/benchmarking.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 10 minutes 3.950 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 6 minutes 51.260 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-tutorial-autotvm-relay-x86-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../_downloads/57a45d9bef1af358191e7d50043e652c/autotvm_relay_x86.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">autotvm_relay_x86.py</span></code></a></p>
diff --git a/docs/tutorial/cross_compilation_and_rpc.html b/docs/tutorial/cross_compilation_and_rpc.html
index 415ba9547..668998545 100644
--- a/docs/tutorial/cross_compilation_and_rpc.html
+++ b/docs/tutorial/cross_compilation_and_rpc.html
@@ -496,7 +496,7 @@ device and returns the measured cost. Network overhead is excluded.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>1.281e-07 secs/op
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>1.239e-07 secs/op
</pre></div>
</div>
</div>
diff --git a/docs/tutorial/intro_topi.html b/docs/tutorial/intro_topi.html
index 15165ad53..182927fe2 100644
--- a/docs/tutorial/intro_topi.html
+++ b/docs/tutorial/intro_topi.html
@@ -461,7 +461,7 @@ we can schedule the following series of operations ending with <code class="code
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[stage(a, placeholder(a, 0x22cf41b0)), stage(b, placeholder(b, 0x2334a8f0)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[ [...]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[stage(a, placeholder(a, 0x207ddba0)), stage(b, placeholder(b, 0x1a5492f0)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[ [...]
</pre></div>
</div>
<p>We can test the correctness by comparing with <code class="code docutils literal notranslate"><span class="pre">numpy</span></code> result as follows</p>
diff --git a/docs/tutorial/sg_execution_times.html b/docs/tutorial/sg_execution_times.html
index c9ab5b104..669bf50e3 100644
--- a/docs/tutorial/sg_execution_times.html
+++ b/docs/tutorial/sg_execution_times.html
@@ -300,20 +300,20 @@
<div class="section" id="computation-times">
<span id="sphx-glr-tutorial-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>13:12.953</strong> total execution time for <strong>tutorial</strong> files:</p>
+<p><strong>09:45.828</strong> total execution time for <strong>tutorial</strong> files:</p>
<ul class="simple">
-<li><p><strong>10:03.950</strong>: <a class="reference internal" href="autotvm_relay_x86.html#sphx-glr-tutorial-autotvm-relay-x86-py"><span class="std std-ref">Compiling and Optimizing a Model with the Python Interface (AutoTVM)</span></a> (<code class="docutils literal notranslate"><span class="pre">autotvm_relay_x86.py</span></code>)</p></li>
-<li><p><strong>01:07.232</strong>: <a class="reference internal" href="auto_scheduler_matmul_x86.html#sphx-glr-tutorial-auto-scheduler-matmul-x86-py"><span class="std std-ref">Optimizing Operators with Auto-scheduling</span></a> (<code class="docutils literal notranslate"><span class="pre">auto_scheduler_matmul_x86.py</span></code>)</p></li>
-<li><p><strong>00:58.830</strong>: <a class="reference internal" href="tensor_expr_get_started.html#sphx-glr-tutorial-tensor-expr-get-started-py"><span class="std std-ref">Working with Operators Using Tensor Expression</span></a> (<code class="docutils literal notranslate"><span class="pre">tensor_expr_get_started.py</span></code>)</p></li>
-<li><p><strong>00:34.159</strong>: <a class="reference internal" href="autotvm_matmul_x86.html#sphx-glr-tutorial-autotvm-matmul-x86-py"><span class="std std-ref">Optimizing Operators with Schedule Templates and AutoTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">autotvm_matmul_x86.py</span></code>)</p></li>
-<li><p><strong>00:26.513</strong>: <a class="reference internal" href="relay_quick_start.html#sphx-glr-tutorial-relay-quick-start-py"><span class="std std-ref">Quick Start Tutorial for Compiling Deep Learning Models</span></a> (<code class="docutils literal notranslate"><span class="pre">relay_quick_start.py</span></code>)</p></li>
-<li><p><strong>00:01.118</strong>: <a class="reference internal" href="tensor_ir_blitz_course.html#sphx-glr-tutorial-tensor-ir-blitz-course-py"><span class="std std-ref">Blitz Course to TensorIR</span></a> (<code class="docutils literal notranslate"><span class="pre">tensor_ir_blitz_course.py</span></code>)</p></li>
-<li><p><strong>00:00.736</strong>: <a class="reference internal" href="intro_topi.html#sphx-glr-tutorial-intro-topi-py"><span class="std std-ref">Introduction to TOPI</span></a> (<code class="docutils literal notranslate"><span class="pre">intro_topi.py</span></code>)</p></li>
-<li><p><strong>00:00.218</strong>: <a class="reference internal" href="cross_compilation_and_rpc.html#sphx-glr-tutorial-cross-compilation-and-rpc-py"><span class="std std-ref">Cross Compilation and RPC</span></a> (<code class="docutils literal notranslate"><span class="pre">cross_compilation_and_rpc.py</span></code>)</p></li>
-<li><p><strong>00:00.053</strong>: <a class="reference internal" href="introduction.html#sphx-glr-tutorial-introduction-py"><span class="std std-ref">Introduction</span></a> (<code class="docutils literal notranslate"><span class="pre">introduction.py</span></code>)</p></li>
-<li><p><strong>00:00.049</strong>: <a class="reference internal" href="tvmc_python.html#sphx-glr-tutorial-tvmc-python-py"><span class="std std-ref">Getting Starting using TVMC Python: a high-level API for TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">tvmc_python.py</span></code>)</p></li>
-<li><p><strong>00:00.047</strong>: <a class="reference internal" href="install.html#sphx-glr-tutorial-install-py"><span class="std std-ref">Installing TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">install.py</span></code>)</p></li>
-<li><p><strong>00:00.047</strong>: <a class="reference internal" href="tvmc_command_line_driver.html#sphx-glr-tutorial-tvmc-command-line-driver-py"><span class="std std-ref">Compiling and Optimizing a Model with TVMC</span></a> (<code class="docutils literal notranslate"><span class="pre">tvmc_command_line_driver.py</span></code>)</p></li>
+<li><p><strong>06:51.260</strong>: <a class="reference internal" href="autotvm_relay_x86.html#sphx-glr-tutorial-autotvm-relay-x86-py"><span class="std std-ref">Compiling and Optimizing a Model with the Python Interface (AutoTVM)</span></a> (<code class="docutils literal notranslate"><span class="pre">autotvm_relay_x86.py</span></code>)</p></li>
+<li><p><strong>01:03.810</strong>: <a class="reference internal" href="auto_scheduler_matmul_x86.html#sphx-glr-tutorial-auto-scheduler-matmul-x86-py"><span class="std std-ref">Optimizing Operators with Auto-scheduling</span></a> (<code class="docutils literal notranslate"><span class="pre">auto_scheduler_matmul_x86.py</span></code>)</p></li>
+<li><p><strong>01:00.772</strong>: <a class="reference internal" href="tensor_expr_get_started.html#sphx-glr-tutorial-tensor-expr-get-started-py"><span class="std std-ref">Working with Operators Using Tensor Expression</span></a> (<code class="docutils literal notranslate"><span class="pre">tensor_expr_get_started.py</span></code>)</p></li>
+<li><p><strong>00:25.543</strong>: <a class="reference internal" href="relay_quick_start.html#sphx-glr-tutorial-relay-quick-start-py"><span class="std std-ref">Quick Start Tutorial for Compiling Deep Learning Models</span></a> (<code class="docutils literal notranslate"><span class="pre">relay_quick_start.py</span></code>)</p></li>
+<li><p><strong>00:22.333</strong>: <a class="reference internal" href="autotvm_matmul_x86.html#sphx-glr-tutorial-autotvm-matmul-x86-py"><span class="std std-ref">Optimizing Operators with Schedule Templates and AutoTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">autotvm_matmul_x86.py</span></code>)</p></li>
+<li><p><strong>00:01.089</strong>: <a class="reference internal" href="tensor_ir_blitz_course.html#sphx-glr-tutorial-tensor-ir-blitz-course-py"><span class="std std-ref">Blitz Course to TensorIR</span></a> (<code class="docutils literal notranslate"><span class="pre">tensor_ir_blitz_course.py</span></code>)</p></li>
+<li><p><strong>00:00.718</strong>: <a class="reference internal" href="intro_topi.html#sphx-glr-tutorial-intro-topi-py"><span class="std std-ref">Introduction to TOPI</span></a> (<code class="docutils literal notranslate"><span class="pre">intro_topi.py</span></code>)</p></li>
+<li><p><strong>00:00.176</strong>: <a class="reference internal" href="cross_compilation_and_rpc.html#sphx-glr-tutorial-cross-compilation-and-rpc-py"><span class="std std-ref">Cross Compilation and RPC</span></a> (<code class="docutils literal notranslate"><span class="pre">cross_compilation_and_rpc.py</span></code>)</p></li>
+<li><p><strong>00:00.034</strong>: <a class="reference internal" href="introduction.html#sphx-glr-tutorial-introduction-py"><span class="std std-ref">Introduction</span></a> (<code class="docutils literal notranslate"><span class="pre">introduction.py</span></code>)</p></li>
+<li><p><strong>00:00.032</strong>: <a class="reference internal" href="tvmc_command_line_driver.html#sphx-glr-tutorial-tvmc-command-line-driver-py"><span class="std std-ref">Compiling and Optimizing a Model with TVMC</span></a> (<code class="docutils literal notranslate"><span class="pre">tvmc_command_line_driver.py</span></code>)</p></li>
+<li><p><strong>00:00.031</strong>: <a class="reference internal" href="install.html#sphx-glr-tutorial-install-py"><span class="std std-ref">Installing TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">install.py</span></code>)</p></li>
+<li><p><strong>00:00.029</strong>: <a class="reference internal" href="tvmc_python.html#sphx-glr-tutorial-tvmc-python-py"><span class="std std-ref">Getting Starting using TVMC Python: a high-level API for TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">tvmc_python.py</span></code>)</p></li>
</ul>
</div>
diff --git a/docs/tutorial/tensor_expr_get_started.html b/docs/tutorial/tensor_expr_get_started.html
index 2a341b9f9..dbf7a3b77 100644
--- a/docs/tutorial/tensor_expr_get_started.html
+++ b/docs/tutorial/tensor_expr_get_started.html
@@ -508,7 +508,7 @@ helper function to run a profile of the TVM generated code.</p>
</div>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.000008
-naive: 0.000008
+naive: 0.000006
</pre></div>
</div>
</div>
@@ -559,7 +559,7 @@ compile and run this new schedule with the parallel operation applied:</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>parallel: 0.000006
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>parallel: 0.000007
</pre></div>
</div>
</div>
@@ -599,7 +599,7 @@ factor to be the number of threads on your CPU.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>vector: 0.000025
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>vector: 0.000026
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [(stride: int32*n: int32)], [], type="auto"),
@@ -633,10 +633,10 @@ factor to be the number of threads on your CPU.</p>
</div>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Operator Timing Performance
- numpy 7.732530000339467e-06 1.0
- naive 7.839e-06 1.0137691026941842
-parallel 6.0116e-06 0.7774428291563155
- vector 2.45931e-05 3.1804726265427146
+ numpy 8.116219999010354e-06 1.0
+ naive 5.8296e-06 0.7182653994976514
+parallel 7.0317e-06 0.866376219577267
+ vector 2.5753399999999996e-05 3.173078108176001
</pre></div>
</div>
<div class="admonition-code-specialization admonition">
@@ -954,7 +954,7 @@ matrix multiplication.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.019003
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.017579
</pre></div>
</div>
<p>Now we write a basic matrix multiplication using TVM TE and verify that it
@@ -996,7 +996,7 @@ optimizations.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>none: 3.211525
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>none: 3.421662
</pre></div>
</div>
<p>Let’s take a look at the intermediate representation of the operator and
@@ -1063,7 +1063,7 @@ schedule.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>blocking: 0.325972
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>blocking: 0.299119
</pre></div>
</div>
<p>By reordering the computation to take advantage of caching, you should see a
@@ -1124,7 +1124,7 @@ already cache friendly from our previous optimizations.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>vectorization: 0.350958
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>vectorization: 0.335124
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1180,7 +1180,7 @@ more cache friendly.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>loop permutation: 0.118166
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>loop permutation: 0.112224
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1257,7 +1257,7 @@ optimized schedule.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>array packing: 0.109252
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>array packing: 0.107586
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1332,7 +1332,7 @@ to `C</cite> when all the block results are ready.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>block caching: 0.110593
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>block caching: 0.110072
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1400,7 +1400,7 @@ of thread-level parallelization.</p>
</pre></div>
</div>
<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>parallelization: 0.145116
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>parallelization: 0.143778
@main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1463,13 +1463,13 @@ working, we can compare the results.</p>
</div>
<p class="sphx-glr-script-out">Out:</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span> Operator Timing Performance
- none 3.2115246612999995 1.0
- blocking 0.325971564 0.10150056386864216
- vectorization 0.3509578926 0.10928077147566884
-loop permutation 0.1181656727 0.03679425978692858
- array packing 0.10925178390000001 0.034018665718660794
- block caching 0.11059276209999999 0.03443621761111837
- parallelization 0.1451155605 0.04518587767632411
+ none 3.4216618965000003 1.0
+ blocking 0.2991194263 0.08741934046901818
+ vectorization 0.3351239071 0.09794185318040817
+loop permutation 0.1122239282 0.0327980763718336
+ array packing 0.1075860693 0.03144263593374004
+ block caching 0.11007208180000001 0.03216918711711177
+ parallelization 0.143777803 0.04201987436194954
</pre></div>
</div>
<p>Note that the outputs on the web page reflect the running times on a
@@ -1501,6 +1501,7 @@ is</p>
you can build generic templates of the matrix multiplication and other
operations with tunable parameters that allows you to automatically optimize
the computation for specific platforms.</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes 0.772 seconds)</p>
<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-tutorial-tensor-expr-get-started-py">
<div class="sphx-glr-download docutils container">
<p><a class="reference download internal" download="" href="../_downloads/40a01cffb015a67aaec0fad7e27cf80d/tensor_expr_get_started.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">tensor_expr_get_started.py</span></code></a></p>