You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Have I written custom code (as opposed to using a stock example script provided in TensorFlow): Yes
OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Ubuntu 14.04
TensorFlow installed from (source or binary): source
TensorFlow version (use command below): 1.2.1
Python version: 3.4.3
Bazel version (if compiling from source): 0.5.2
CUDA/cuDNN version: 8.0/6.0
GPU model and memory: Tesla P100 (16 gb)
Describe the problem
Batch norm layer fails with an error when both center=False, data_format='NCHW' and zero_debias_moving_mean=True arguments are used. It looks like the solution would be just adding additional if to check if beta is None in the same way it is done for gamma, but maybe there are some more dependencies.
Source code / logs
The following code could be used to reproduce the issue:
The Changes have been made as suggested in my previous pull request.
tensorflow#13829
The unit test case has been updated along with the update in layers.py with reference to the issue tensorflow#11673.
Please verify and get back.
System information
Describe the problem
Batch norm layer fails with an error when both center=False, data_format='NCHW' and zero_debias_moving_mean=True arguments are used. It looks like the solution would be just adding additional if to check if beta is None in the same way it is done for gamma, but maybe there are some more dependencies.
Source code / logs
The following code could be used to reproduce the issue:
The error log:
The text was updated successfully, but these errors were encountered: