experiments
/
STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Simplify the padding size calculation 

PiperOrigin-RevId: 314857566
Change-Id: I64a7fd55406d3b072f90c1150d5df0c17cee6a5f
This commit is contained in:
Zhenyu Tan 2020-06-04 20:48:52 -07:00 committed by TensorFlower Gardener
parent 0418b68c73
commit f19c6efb4a
1 changed files with 1 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
tensorflow/python/ops/nn_ops.py
View File

@ -727,10 +727,7 @@ def _with_space_to_batch_base_paddings(filter_shape, num_spatial_dims,
# Spatial dimensions of the filters and the upsampled filters in which we
# introduce (rate - 1) zeros between consecutive filter values.
filter_spatial_shape = filter_shape[:num_spatial_dims]
dilated_filter_spatial_shape = (
filter_spatial_shape + (filter_spatial_shape - 1) *
(rate_or_const_rate - 1))
pad_extra_shape = dilated_filter_spatial_shape - 1
pad_extra_shape = (filter_spatial_shape - 1) * rate_or_const_rate
# When full_padding_shape is odd, we pad more at end, following the same
# convention as conv2d.