optima2005 commented on a change in pull request #4484: [FRONTEND][TF]
conv2d_transpose 'SAME' support kernel more than 1x1
URL: https://github.com/apache/incubator-tvm/pull/4484#discussion_r355807286
##########
File path: tests/python/frontend/tensorflow/test_forward.py
##########
@@ -362,10 +362,16 @@ def test_forward_convolution():
_test_convolution('depthwise', [4, 12, 17, 17], [3, 3, 12, 2], [1, 1],
[2, 2], 'VALID', 'NCHW')
_test_convolution('conv_transpose', [4, 32, 8, 8], [1, 1, 176, 32],
[1, 1], [1, 1], 'SAME',
'NCHW', [4, 176, 8, 8])
+ _test_convolution('conv_transpose', [4, 32, 8, 8], [2, 2, 176, 32],
[1, 1], [1, 1], 'SAME',
+ 'NCHW', [4, 176, 8, 8])
+ _test_convolution('conv_transpose', [4, 32, 8, 8], [3, 3, 176, 32],
[1, 1], [1, 1], 'SAME',
+ 'NCHW', [4, 176, 8, 8])
_test_convolution('conv_transpose', [4, 19, 8, 8], [3, 3, 19, 19], [1,
1], [2, 2], 'VALID',
'NCHW', [4, 19, 17, 17])
_test_convolution('conv_transpose', [4, 19, 17, 17], [1, 1, 124, 19],
[1, 1], [1, 1], 'SAME',
'NCHW', [4, 124, 17, 17])
+ _test_convolution('conv_transpose', [4, 19, 17, 17], [3, 3, 124, 19],
[1, 1], [1, 1], 'SAME',
Review comment:
The strides in transpose convolution is used to dilate the input, see
[this](https://github.com/apache/incubator-tvm/blob/203d2c922dd09298eedf43ae6b0224c60359b169/topi/python/topi/nn/conv2d_transpose.py#L64).
If the input is dilated, there would be no way to get 'SAME' size of output
by only padding.
----------------------------------------------------------------
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:
[email protected]
With regards,
Apache Git Services
