You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by ib...@apache.org on 2019/03/04 01:58:45 UTC
[incubator-mxnet] branch ib/jl-runtime-features updated (93f8e59 ->
6d9e429)
This is an automated email from the ASF dual-hosted git repository.
iblis pushed a change to branch ib/jl-runtime-features
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
discard 93f8e59 update
discard 189478d mx.isenabled
discard 64a2a3d 2 space
discard 3c2132a update to new API
discard d465c52 Julia: add binding for runtime feature detection
new 5f32f32 Dual stream cudnn Convolution backward() with MXNET_GPU_WORKER_NSTREAMS=2. (#14006)
new 0af40f7 [MXNET-1325] Make InferShapeAttr a standalone pass (#14193)
new 0eed3da Temporarily disables windows pipeline to unblock PRs (#14261)
new e3a51b5 [op] add back support for scalar type rescale_grad argument for adamw_update/mp_adamw_update (#14221)
new 7c617cc pypi package description. manifest/setup.py update (#14255)
new f0fedec Refactors USE_NVRTC setting to ENABLE_CUDA_RTC in pip make config files (#14250)
new c319ae5 MXNet Java bug fixes and experience improvement (#14213)
new 992c3c0 [MXNET-1330] Bring nnvm::Tuple to mxnet::Tuple (#14270)
new c6b1fd5 MXNET-1302 Exclude commons-codec and commons-io from assembled JAR (#14000)
new 3180f9c added mkldnn dependency for plugin compile target (#14274)
new fb4f9d5 Large array support for randint (#14242)
new 7aac0b5 reverting broadcasting fixes (#14299)
new 0e23a18 add symbolic link to mkldnn header files in include (#14300)
new 053ffc7 fix memory-related issues to enable ASAN tests (#14223)
new 7b1727a Julia: add binding for runtime feature detection
new ec80b31 update to new API
new 03698eb 2 space
new 271fed6 mx.isenabled
new 6d9e429 update
This update added new revisions after undoing existing revisions.
That is to say, some revisions that were in the old version of the
branch are not in the new version. This situation occurs
when a user --force pushes a change and generates a repository
containing something like this:
* -- * -- B -- O -- O -- O (93f8e59)
\
N -- N -- N refs/heads/ib/jl-runtime-features (6d9e429)
You should already have received notification emails for all of the O
revisions, and so the following emails describe only the N revisions
from the common base, B.
Any revisions marked "omit" are not gone; other references still
refer to them. Any revisions marked "discard" are gone forever.
The 9336 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "add" were already present in the repository and have only
been added to this reference.
Summary of changes:
3rdparty/dmlc-core | 2 +-
3rdparty/mshadow | 2 +-
CONTRIBUTORS.md | 1 +
Makefile | 2 +-
amalgamation/prep_nnvm.sh | 1 -
ci/docker/runtime_functions.sh | 2 -
ci/jenkins/Jenkinsfile_windows_cpu | 10 +-
ci/jenkins/Jenkinsfile_windows_gpu | 10 +-
cpp-package/example/alexnet.cpp | 1 +
cpp-package/example/charRNN.cpp | 10 +
cpp-package/example/googlenet.cpp | 1 +
cpp-package/example/inception_bn.cpp | 1 +
cpp-package/example/lenet.cpp | 1 +
cpp-package/example/lenet_with_mxdataiter.cpp | 1 +
cpp-package/example/mlp_cpu.cpp | 1 +
cpp-package/example/mlp_csv.cpp | 1 +
cpp-package/example/mlp_gpu.cpp | 1 +
cpp-package/example/resnet.cpp | 1 +
cpp-package/example/test_optimizer.cpp | 1 +
cpp-package/example/test_score.cpp | 1 +
docs/architecture/overview.md | 22 +-
docs/faq/add_op_in_backend.md | 8 +-
docs/faq/env_var.md | 6 +
docs/faq/new_op.md | 2 +-
.../predict-cpp/image-classification-predict.cc | 4 +-
include/mkldnn | 1 +
include/mxnet/base.h | 137 ++++-
include/mxnet/executor.h | 4 +-
include/mxnet/ndarray.h | 93 +--
include/mxnet/op_attr_types.h | 11 +-
include/mxnet/operator.h | 20 +-
include/mxnet/operator_util.h | 8 +-
include/mxnet/tensor_blob.h | 12 +-
include/mxnet/tuple.h | 682 +++++++++++++++++++++
make/pip/pip_darwin_cpu.mk | 2 +-
make/pip/pip_darwin_mkl.mk | 2 +-
make/pip/pip_linux_cpu.mk | 2 +-
make/pip/pip_linux_cu100.mk | 2 +-
make/pip/pip_linux_cu100mkl.mk | 2 +-
make/pip/pip_linux_cu75.mk | 2 +-
make/pip/pip_linux_cu75mkl.mk | 2 +-
make/pip/pip_linux_cu80.mk | 2 +-
make/pip/pip_linux_cu80mkl.mk | 2 +-
make/pip/pip_linux_cu90.mk | 2 +-
make/pip/pip_linux_cu90mkl.mk | 2 +-
make/pip/pip_linux_cu91.mk | 2 +-
make/pip/pip_linux_cu91mkl.mk | 2 +-
make/pip/pip_linux_cu92.mk | 2 +-
make/pip/pip_linux_cu92mkl.mk | 2 +-
make/pip/pip_linux_mkl.mk | 2 +-
plugin/caffe/caffe_blob.cc | 4 +-
plugin/caffe/caffe_blob.h | 4 +-
plugin/caffe/caffe_loss-inl.h | 12 +-
plugin/caffe/caffe_loss.cc | 4 +-
plugin/caffe/caffe_op-inl.h | 14 +-
plugin/caffe/caffe_op.cc | 4 +-
plugin/opencv/cv_api.cc | 6 +-
plugin/sframe/iter_sframe.cc | 4 +-
plugin/torch/torch_criterion-inl.h | 14 +-
plugin/torch/torch_module-inl.h | 12 +-
plugin/warpctc/warpctc-inl.h | 12 +-
.../mxnet/contrib/onnx/onnx2mx/_op_translations.py | 36 +-
.../contrib/onnx/onnx2mx/_translation_utils.py | 14 -
python/mxnet/ndarray/contrib.py | 26 +
python/mxnet/ndarray/register.py | 11 +
python/mxnet/symbol/contrib.py | 22 +
python/mxnet/symbol/register.py | 11 +
.../assembly/src/main/assembly/assembly.xml | 2 +
scala-package/core/pom.xml | 5 -
.../scala/org/apache/mxnet/javaapi/NDArray.scala | 2 +
scala-package/deploy/src/main/deploy/deploy.xml | 10 +
scala-package/macros/pom.xml | 5 -
scala-package/mxnet-demo/java-demo/README.md | 18 +-
.../mxnet-demo/java-demo/bin/java_sample.sh | 2 +-
scala-package/mxnet-demo/java-demo/bin/run_od.sh | 2 +-
scala-package/mxnet-demo/java-demo/pom.xml | 6 +
.../src/main/java/mxnet/ImageClassification.java} | 95 +--
.../{HelloWorld.java => NDArrayCreation.java} | 29 +-
.../{HelloWorld.java => NDArrayOperation.java} | 26 +-
.../src/main/java/mxnet/ObjectDetection.java | 16 +-
scala-package/pom.xml | 5 +
src/c_api/c_api.cc | 18 +-
src/c_api/c_api_common.h | 4 +-
src/c_api/c_api_executor.cc | 8 +-
src/c_api/c_api_function.cc | 10 +-
src/c_api/c_api_symbolic.cc | 10 +-
src/c_api/c_predict_api.cc | 42 +-
src/common/exec_utils.h | 8 +-
src/common/object_pool.h | 7 +-
src/common/serialization.h | 1 -
src/common/utils.h | 14 +-
src/engine/naive_engine.cc | 25 +-
src/engine/stream_manager.h | 25 +-
src/engine/threaded_engine.cc | 2 +-
src/engine/threaded_engine.h | 3 +-
src/engine/threaded_engine_perdevice.cc | 12 +-
src/executor/attach_op_execs_pass.cc | 6 +-
src/executor/exec_pass.h | 53 +-
src/executor/graph_executor.cc | 59 +-
src/executor/graph_executor.h | 8 +-
src/executor/infer_graph_attr_pass.cc | 274 ++++++++-
src/executor/tensorrt_pass.cc | 8 +-
src/executor/trt_graph_executor.cc | 21 +-
src/executor/trt_graph_executor.h | 8 +-
src/imperative/cached_op.cc | 20 +-
src/imperative/imperative.cc | 4 +-
src/imperative/imperative_utils.h | 26 +-
src/io/image_aug_default.cc | 2 +-
src/io/image_det_aug_default.cc | 2 +-
src/io/image_io.cc | 17 +-
src/io/image_iter_common.h | 2 +-
src/io/inst_vector.h | 2 +-
src/io/iter_batchloader.h | 6 +-
src/io/iter_csv.cc | 8 +-
src/io/iter_image_det_recordio.cc | 2 +-
src/io/iter_image_recordio_2.cc | 4 +-
src/io/iter_libsvm.cc | 12 +-
src/io/iter_mnist.cc | 2 +-
src/io/iter_sparse.h | 2 +-
src/io/iter_sparse_batchloader.h | 8 +-
src/io/iter_sparse_prefetcher.h | 2 +-
src/kvstore/comm.h | 14 +-
src/kvstore/comm_tree.h | 8 +-
src/kvstore/kvstore_dist.h | 5 +-
src/kvstore/kvstore_dist_server.h | 10 +-
src/kvstore/kvstore_nccl.h | 6 +-
src/ndarray/ndarray.cc | 66 +-
src/ndarray/ndarray_function.h | 10 +-
src/nnvm/gradient.cc | 281 +++++++++
src/nnvm/graph_algorithm.h | 131 ++++
src/nnvm/legacy_op_util.cc | 20 +-
src/nnvm/plan_memory.cc | 412 +++++++++++++
src/operator/batch_norm_v1-inl.h | 16 +-
src/operator/batch_norm_v1.cc | 4 +-
src/operator/bilinear_sampler-inl.h | 12 +-
src/operator/bilinear_sampler.cc | 2 +-
src/operator/contrib/adamw-inl.h | 8 +-
src/operator/contrib/adamw.cc | 14 +-
src/operator/contrib/adamw.cu | 4 +-
src/operator/contrib/adaptive_avg_pooling-inl.h | 10 +-
src/operator/contrib/adaptive_avg_pooling.cc | 2 +-
src/operator/contrib/bilinear_resize-inl.h | 6 +-
src/operator/contrib/bilinear_resize.cc | 2 +-
src/operator/contrib/boolean_mask.cc | 2 +-
src/operator/contrib/boolean_mask.cu | 2 +-
src/operator/contrib/bounding_box-inl.h | 37 +-
src/operator/contrib/bounding_box.cc | 6 +-
src/operator/contrib/count_sketch-inl.h | 18 +-
src/operator/contrib/count_sketch.cc | 4 +-
src/operator/contrib/deformable_convolution-inl.h | 44 +-
src/operator/contrib/deformable_convolution.cc | 8 +-
src/operator/contrib/deformable_convolution.cu | 4 +-
.../contrib/deformable_psroi_pooling-inl.h | 16 +-
src/operator/contrib/deformable_psroi_pooling.cc | 4 +-
src/operator/contrib/dgl_graph.cc | 68 +-
src/operator/contrib/fft-inl.h | 24 +-
src/operator/contrib/fft.cc | 2 +-
src/operator/contrib/ifft-inl.h | 22 +-
src/operator/contrib/ifft.cc | 2 +-
src/operator/contrib/index_copy-inl.h | 4 +-
src/operator/contrib/index_copy.cc | 2 +-
src/operator/contrib/krprod.cc | 8 +-
src/operator/contrib/multi_proposal-inl.h | 10 +-
src/operator/contrib/multibox_detection-inl.h | 21 +-
src/operator/contrib/multibox_detection.cc | 4 +-
src/operator/contrib/multibox_prior-inl.h | 13 +-
src/operator/contrib/multibox_prior.cc | 4 +-
src/operator/contrib/multibox_target-inl.h | 23 +-
src/operator/contrib/multibox_target.cc | 4 +-
src/operator/contrib/nn/deformable_im2col.cuh | 18 +-
src/operator/contrib/nn/deformable_im2col.h | 20 +-
src/operator/contrib/nnvm_to_onnx-inl.h | 6 +-
src/operator/contrib/nnvm_to_onnx.cc | 30 +-
src/operator/contrib/nnz.cc | 6 +-
src/operator/contrib/optimizer_op.cc | 6 +-
src/operator/contrib/proposal-inl.h | 10 +-
src/operator/contrib/psroi_pooling-inl.h | 14 +-
src/operator/contrib/psroi_pooling.cc | 4 +-
src/operator/contrib/quadratic_op-inl.h | 4 +-
src/operator/contrib/quadratic_op.cc | 2 +-
src/operator/contrib/roi_align-inl.h | 2 +-
src/operator/contrib/roi_align.cc | 8 +-
src/operator/contrib/sync_batch_norm-inl.h | 16 +-
src/operator/contrib/sync_batch_norm.cc | 4 +-
src/operator/contrib/tensorrt.cc | 8 +-
src/operator/control_flow.cc | 68 +-
src/operator/convolution_v1-inl.h | 34 +-
src/operator/convolution_v1.cc | 8 +-
src/operator/convolution_v1.cu | 4 +-
src/operator/correlation-inl.h | 12 +-
src/operator/correlation.cc | 2 +-
src/operator/crop-inl.h | 18 +-
src/operator/cross_device_copy.cc | 8 +-
src/operator/custom/custom.cc | 14 +-
src/operator/custom/native_op-inl.h | 16 +-
src/operator/custom/ndarray_op-inl.h | 10 +-
src/operator/elemwise_op_common.h | 8 +-
src/operator/grid_generator-inl.h | 22 +-
src/operator/grid_generator.cc | 2 +-
src/operator/identity_attach_KL_sparse_reg-inl.h | 10 +-
src/operator/image/image_random-inl.h | 24 +-
src/operator/image/image_random.cc | 4 +-
src/operator/image/resize-inl.h | 8 +-
src/operator/image/resize.cc | 2 +-
src/operator/instance_norm-inl.h | 14 +-
src/operator/instance_norm.cc | 2 +-
src/operator/l2_normalization-inl.h | 18 +-
src/operator/l2_normalization.cc | 4 +-
src/operator/leaky_relu-inl.h | 30 +-
src/operator/leaky_relu.cc | 2 +-
src/operator/loss_binary_op-inl.h | 6 +-
src/operator/loss_binary_op.cc | 2 +-
src/operator/make_loss-inl.h | 12 +-
src/operator/make_loss.cc | 4 +-
src/operator/nn/activation.cc | 2 +-
src/operator/nn/batch_norm-inl.h | 4 +-
src/operator/nn/batch_norm.cc | 20 +-
src/operator/nn/batch_norm.cu | 4 +-
src/operator/nn/concat.cc | 24 +-
src/operator/nn/convolution-inl.h | 20 +-
src/operator/nn/convolution.cc | 10 +-
src/operator/nn/convolution.cu | 20 +-
src/operator/nn/ctc_loss-inl.h | 14 +-
src/operator/nn/ctc_loss.cc | 2 +-
src/operator/nn/cudnn/cudnn_algoreg-inl.h | 6 +-
src/operator/nn/cudnn/cudnn_batch_norm.cc | 16 +-
src/operator/nn/cudnn/cudnn_convolution-inl.h | 127 ++--
src/operator/nn/cudnn/cudnn_deconvolution-inl.h | 56 +-
src/operator/nn/deconvolution-inl.h | 40 +-
src/operator/nn/deconvolution.cc | 10 +-
src/operator/nn/deconvolution.cu | 12 +-
src/operator/nn/depthwise_convolution-inl.h | 4 +-
src/operator/nn/dropout-inl.h | 12 +-
src/operator/nn/dropout.cc | 6 +-
src/operator/nn/fully_connected-inl.h | 8 +-
src/operator/nn/fully_connected.cc | 12 +-
src/operator/nn/im2col.cuh | 16 +-
src/operator/nn/im2col.h | 22 +-
src/operator/nn/layer_norm-inl.h | 8 +-
src/operator/nn/layer_norm.cc | 14 +-
src/operator/nn/lrn.cc | 8 +-
src/operator/nn/mkldnn/mkldnn_base-inl.h | 10 +-
src/operator/nn/mkldnn/mkldnn_fully_connected.cc | 8 +-
src/operator/nn/mkldnn/mkldnn_pooling-inl.h | 2 +-
src/operator/nn/mkldnn/mkldnn_slice.cc | 4 +-
src/operator/nn/pool.cuh | 44 +-
src/operator/nn/pool.h | 160 ++---
src/operator/nn/pooling-inl.h | 44 +-
src/operator/nn/pooling.cc | 16 +-
src/operator/nn/softmax-inl.h | 8 +-
src/operator/nn/softmax.cc | 12 +-
src/operator/nn/upsampling-inl.h | 8 +-
src/operator/nn/upsampling.cc | 8 +-
src/operator/nnpack/nnpack_fully_connected-inl.h | 4 +-
src/operator/operator_common.h | 14 +-
src/operator/operator_util.cc | 34 +-
src/operator/optimizer_op-inl.h | 8 +-
src/operator/optimizer_op.cc | 32 +-
src/operator/pad-inl.h | 12 +-
src/operator/pad.cc | 2 +-
src/operator/pooling_v1-inl.h | 30 +-
src/operator/pooling_v1.cc | 4 +-
src/operator/quantization/dequantize-inl.h | 6 +-
src/operator/quantization/dequantize.cc | 2 +-
.../quantization/mkldnn/mkldnn_requantize-inl.h | 4 +-
src/operator/quantization/quantization_utils.h | 8 +-
src/operator/quantization/quantize-inl.h | 10 +-
src/operator/quantization/quantize.cc | 2 +-
src/operator/quantization/quantize_v2-inl.h | 13 +-
src/operator/quantization/quantize_v2.cc | 2 +-
src/operator/quantization/quantized_concat.cc | 18 +-
src/operator/quantization/quantized_conv.cc | 18 +-
src/operator/quantization/quantized_conv.cu | 34 +-
src/operator/quantization/quantized_flatten-inl.h | 14 +-
src/operator/quantization/quantized_flatten.cc | 2 +-
.../quantization/quantized_fully_connected.cc | 26 +-
.../quantization/quantized_fully_connected.cu | 6 +-
src/operator/quantization/quantized_pooling.cc | 18 +-
src/operator/quantization/quantized_pooling.cu | 2 +-
src/operator/quantization/requantize-inl.h | 4 +-
src/operator/quantization/requantize.cc | 2 +-
src/operator/random/multisample_op.cc | 2 +-
src/operator/random/multisample_op.h | 16 +-
src/operator/random/sample_multinomial_op.cc | 2 +-
src/operator/random/sample_multinomial_op.h | 16 +-
src/operator/random/sample_op.cc | 4 +-
src/operator/random/sample_op.h | 18 +-
src/operator/random/sampler.h | 4 +-
src/operator/random/shuffle_op.cc | 4 +-
src/operator/random/shuffle_op.cu | 2 +-
src/operator/random/unique_sample_op.cc | 2 +-
src/operator/random/unique_sample_op.h | 8 +-
src/operator/regression_output-inl.h | 8 +-
src/operator/regression_output.cc | 2 +-
src/operator/rnn-inl.h | 20 +-
src/operator/rnn.cc | 2 +-
src/operator/roi_pooling-inl.h | 14 +-
src/operator/roi_pooling.cc | 2 +-
src/operator/sequence_last-inl.h | 16 +-
src/operator/sequence_last.cc | 2 +-
src/operator/sequence_mask-inl.h | 12 +-
src/operator/sequence_mask.cc | 2 +-
src/operator/sequence_reverse-inl.h | 12 +-
src/operator/sequence_reverse.cc | 2 +-
src/operator/slice_channel-inl.h | 16 +-
src/operator/slice_channel.cc | 2 +-
src/operator/softmax_output-inl.h | 20 +-
src/operator/softmax_output.cc | 16 +-
src/operator/spatial_transformer-inl.h | 20 +-
src/operator/spatial_transformer.cc | 2 +-
src/operator/subgraph/common.h | 12 +-
src/operator/subgraph/mkldnn/mkldnn_conv.cc | 14 +-
src/operator/subgraph_op_common.cc | 10 +-
src/operator/subgraph_op_common.h | 6 +-
src/operator/svm_output-inl.h | 16 +-
src/operator/svm_output.cc | 2 +-
src/operator/swapaxis-inl.h | 18 +-
src/operator/swapaxis.cc | 2 +-
src/operator/tensor/broadcast_reduce-inl.cuh | 12 +-
src/operator/tensor/broadcast_reduce-inl.h | 9 +-
src/operator/tensor/broadcast_reduce_op.h | 164 ++---
src/operator/tensor/broadcast_reduce_op_index.cc | 4 +-
src/operator/tensor/broadcast_reduce_op_value.cc | 10 +-
src/operator/tensor/broadcast_reduce_op_value.cu | 2 +-
src/operator/tensor/cast_storage-inl.cuh | 1 -
src/operator/tensor/cast_storage-inl.h | 2 +-
src/operator/tensor/cast_storage.cc | 2 +-
src/operator/tensor/control_flow_op.cc | 2 +-
src/operator/tensor/control_flow_op.h | 6 +-
src/operator/tensor/diag_op-inl.h | 26 +-
src/operator/tensor/diag_op.cc | 2 +-
src/operator/tensor/dot-inl.h | 32 +-
src/operator/tensor/dot.cc | 4 +-
.../tensor/elemwise_binary_broadcast_op-inl.cuh | 2 +-
src/operator/tensor/elemwise_binary_broadcast_op.h | 44 +-
src/operator/tensor/elemwise_binary_op-inl.h | 2 +-
src/operator/tensor/elemwise_binary_op.h | 2 +-
src/operator/tensor/elemwise_binary_scalar_op.h | 2 +-
.../tensor/elemwise_binary_scalar_op_basic.cc | 2 +-
.../tensor/elemwise_binary_scalar_op_extended.cc | 2 +-
src/operator/tensor/elemwise_sum.cc | 10 +-
src/operator/tensor/elemwise_unary_op.h | 10 +-
src/operator/tensor/elemwise_unary_op_basic.cc | 34 +-
src/operator/tensor/histogram-inl.h | 12 +-
src/operator/tensor/histogram.cc | 2 +-
src/operator/tensor/indexing_op.cc | 34 +-
src/operator/tensor/indexing_op.cu | 10 +-
src/operator/tensor/indexing_op.h | 86 +--
src/operator/tensor/init_op.cc | 16 +-
src/operator/tensor/init_op.h | 28 +-
src/operator/tensor/la_op.cc | 20 +-
src/operator/tensor/la_op.h | 56 +-
src/operator/tensor/matrix_op-inl.h | 281 ++++-----
src/operator/tensor/matrix_op.cc | 42 +-
src/operator/tensor/matrix_op.cu | 4 +-
src/operator/tensor/ordering_op-inl.h | 38 +-
src/operator/tensor/ordering_op.cc | 6 +-
src/operator/tensor/ravel.cc | 4 +-
src/operator/tensor/ravel.h | 20 +-
src/operator/tensor/sparse_retain-inl.h | 6 +-
src/operator/tensor/sparse_retain.cc | 2 +-
src/operator/tensor/square_sum-inl.h | 6 +-
src/profiler/profiler.h | 4 +-
tests/cpp/include/test_core_op.h | 22 +-
tests/cpp/include/test_legacy_op.h | 16 +-
tests/cpp/include/test_mkldnn.h | 46 +-
tests/cpp/include/test_ndarray_utils.h | 20 +-
tests/cpp/include/test_op.h | 4 +-
tests/cpp/include/test_op_runner.h | 12 +-
tests/cpp/include/test_tune.h | 8 +-
tests/cpp/include/test_util.h | 44 +-
tests/cpp/misc/serialization.cc | 10 +-
tests/cpp/operator/activation_perf.cc | 14 +-
tests/cpp/operator/batchnorm_test.cc | 48 +-
tests/cpp/operator/coreop_perf.cc | 6 +-
tests/cpp/operator/dropout_perf.cc | 14 +-
tests/cpp/operator/fully_conn_perf.cc | 25 +-
tests/cpp/operator/mkldnn_operator_test.cc | 30 +-
tests/cpp/operator/mkldnn_test.cc | 4 +-
tests/cpp/operator/runner/core_op_runner_test.cc | 44 +-
tests/cpp/operator/slice_channel_perf.cc | 14 +-
tests/cpp/operator/tune/operator_tune_test.cc | 6 +-
tests/nightly/test_large_array.py | 14 +
tests/python/gpu/test_operator_gpu.py | 50 ++
tests/python/unittest/test_contrib_optimizer.py | 12 +
tools/pip/MANIFEST.in | 3 +
tools/pip/doc/CPU_ADDITIONAL.md | 40 ++
tools/pip/doc/CU100MKL_ADDITIONAL.md | 44 ++
tools/pip/doc/CU100_ADDITIONAL.md | 44 ++
tools/pip/doc/CU75MKL_ADDITIONAL.md | 42 ++
tools/pip/doc/CU75_ADDITIONAL.md | 42 ++
tools/pip/doc/CU80MKL_ADDITIONAL.md | 42 ++
tools/pip/doc/CU80_ADDITIONAL.md | 42 ++
tools/pip/doc/CU90MKL_ADDITIONAL.md | 42 ++
tools/pip/doc/CU90_ADDITIONAL.md | 42 ++
tools/pip/doc/CU91MKL_ADDITIONAL.md | 42 ++
tools/pip/doc/CU91_ADDITIONAL.md | 42 ++
tools/pip/doc/CU92MKL_ADDITIONAL.md | 42 ++
tools/pip/doc/CU92_ADDITIONAL.md | 42 ++
tools/pip/doc/MKL_ADDITIONAL.md | 40 ++
docs/api/index.md => tools/pip/doc/PYPI_README.md | 18 +-
tools/pip/setup.py | 17 +-
402 files changed, 5374 insertions(+), 2458 deletions(-)
create mode 120000 include/mkldnn
create mode 100644 include/mxnet/tuple.h
copy scala-package/{examples/src/main/java/org/apache/mxnetexamples/javaapi/infer/predictor/PredictorExample.java => mxnet-demo/java-demo/src/main/java/mxnet/ImageClassification.java} (60%)
copy scala-package/mxnet-demo/java-demo/src/main/java/mxnet/{HelloWorld.java => NDArrayCreation.java} (59%)
rename scala-package/mxnet-demo/java-demo/src/main/java/mxnet/{HelloWorld.java => NDArrayOperation.java} (67%)
create mode 100644 src/nnvm/gradient.cc
create mode 100644 src/nnvm/graph_algorithm.h
create mode 100644 src/nnvm/plan_memory.cc
create mode 100644 tools/pip/doc/CPU_ADDITIONAL.md
create mode 100644 tools/pip/doc/CU100MKL_ADDITIONAL.md
create mode 100644 tools/pip/doc/CU100_ADDITIONAL.md
create mode 100644 tools/pip/doc/CU75MKL_ADDITIONAL.md
create mode 100644 tools/pip/doc/CU75_ADDITIONAL.md
create mode 100644 tools/pip/doc/CU80MKL_ADDITIONAL.md
create mode 100644 tools/pip/doc/CU80_ADDITIONAL.md
create mode 100644 tools/pip/doc/CU90MKL_ADDITIONAL.md
create mode 100644 tools/pip/doc/CU90_ADDITIONAL.md
create mode 100644 tools/pip/doc/CU91MKL_ADDITIONAL.md
create mode 100644 tools/pip/doc/CU91_ADDITIONAL.md
create mode 100644 tools/pip/doc/CU92MKL_ADDITIONAL.md
create mode 100644 tools/pip/doc/CU92_ADDITIONAL.md
create mode 100644 tools/pip/doc/MKL_ADDITIONAL.md
copy docs/api/index.md => tools/pip/doc/PYPI_README.md (65%)