Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fix issue for float16 data type with reuse in CudnnLSTM #18674

Merged
merged 2 commits into from
Apr 19, 2018
Merged

Fix issue for float16 data type with reuse in CudnnLSTM #18674

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
ac47177
Fix issue for float16 data type with reuse in CudnnLSTM
yongtang Apr 18, 2018
3836be5
Fix pylint `Line too long (102/80)`
yongtang Apr 18, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
3 changes: 2 additions & 1 deletion tensorflow/contrib/cudnn_rnn/python/layers/cudnn_rnn.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -358,7 +358,8 @@ def build(self, input_shape):
"CUDA/CuDNN generations.")
# Initialize opaque params with a tensor.
self.kernel = vs.get_variable(
"opaque_kernel", initializer=opaque_params_t, validate_shape=False)
"opaque_kernel", dtype=self._plain_dtype,
initializer=opaque_params_t, validate_shape=False)
# Create saveable in the outer scope of the cudnn subgraph, such that
# alternative subgraph with platform-independent rnn cells can load the
# checkpoints directly.
Expand Down