You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2020/06/23 19:44:55 UTC

[GitHub] [incubator-tvm] electriclilies commented on a change in pull request #5826: [DYNAMIC] Add Dynamic reshape to a dynamic namespace and add DynamicToStatic Pass

electriclilies commented on a change in pull request #5826:
URL: https://github.com/apache/incubator-tvm/pull/5826#discussion_r444465147



##########
File path: python/tvm/relay/op/dyn/transform.py
##########
@@ -0,0 +1,74 @@
+# 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.
+
+# pylint: disable=import-outside-toplevel
+"""Dynamic Transform operators."""
+
+from . import _make
+
+
+def reshape(data, newshape):
+    """Reshape the input array based on the values of the newshape tensor.
+
+    To give user more convenience in without doing manual shape inference,
+    some dimensions of the shape can take special values from the set {0, -1, -3}.
+    The significance of each is explained below:
+
+    ``0`` copy this dimension from the input to the output shape.
+
+        .. code-block:: python
+
+            data.shape = (2,3,4), newshape = (4,0,2), result.shape = (4,3,2)
+            data.shape = (2,3,4), newshape = (2,0,0), result.shape = (2,3,4)
+
+    ``-1`` infers the dimension of the output shape by using the remainder of
+    the input dimensions keeping the size of the new array same as that of the input array.
+    At most one dimension of shape can be -1.
+
+        .. code-block:: python
+
+            data.shape = (2,3,4), newshape = (6,1,-1), result.shape = (6,1,4)
+            data.shape = (2,3,4), newshape = (3,-1,8), result.shape = (3,1,8)
+            data.shape = (2,3,4), newshape = (-1,), result.shape = (24,)
+
+    ``-3`` use the product of two consecutive dimensions of the input shape
+    as the output dimension.
+
+        .. code-block:: python
+
+            data.shape = (2,3,4), newshape = (-3,4), result.shape = (6,4)
+            data.shape = (2,3,4,5), newshape = (-3,-3), result.shape = (6,20)
+            data.shape = (2,3,4), newshape = (0,-3), result.shape = (2,12)
+            data.shape = (2,3,4), newshape = (-3,-2), result.shape = (6,4)

Review comment:
       This example uses -2 as an argument (in the last example), but it was noted earlier that -2 is not allowed as an argument in dynamic reshape. 




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org