[tvm] branch ci-docker-staging updated: support explicit padding for NCHW TF padding test

[mbrookhart]

This is an automated email from the ASF dual-hosted git repository.

mbrookhart pushed a commit to branch ci-docker-staging
in repository https://gitbox.apache.org/repos/asf/tvm.git


The following commit(s) were added to refs/heads/ci-docker-staging by this push:
     new c6064b0  support explicit padding for NCHW TF padding test
c6064b0 is described below

commit c6064b041845bd3b49a6c724804f54c2954d1179
Author: Matthew Brookhart <mbrookh...@octoml.ai>
AuthorDate: Wed Jul 14 10:14:04 2021 -0700

    support explicit padding for NCHW TF padding test
---
 tests/python/frontend/tensorflow/test_forward.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/python/frontend/tensorflow/test_forward.py 
b/tests/python/frontend/tensorflow/test_forward.py
index 7cd47d0..72af268 100644
--- a/tests/python/frontend/tensorflow/test_forward.py
+++ b/tests/python/frontend/tensorflow/test_forward.py
@@ -325,6 +325,8 @@ def _test_pooling(input_shape, **kwargs):
     if is_gpu_available():
         if len(input_shape) == 4:
             input_shape = [input_shape[ii] for ii in (0, 3, 1, 2)]
+            if isinstance(kwargs["padding"], list):
+                kwargs["padding"] = [kwargs["padding"][ii] for ii in (0, 3, 1, 
2)]
             kwargs["data_format"] = "NCHW"
             _test_pooling_iteration(input_shape, **kwargs)