Skip to content

GitLab

Commit 6d1dd03d authored by Hongkun Yu's avatar Hongkun Yu Committed by A. Unique TensorFlower
Browse files

Remove vars dedup as keras fixed it. 

PiperOrigin-RevId: 267525663
parent 32974988
Hide whitespace changes
Inline Side-by-side
Showing with 2 additions and 3 deletions
official/bert/model_training_utils.py
View file @ 6d1dd03d
...@@ -231,9 +231,8 @@ def run_customized_training_loop( ...@@ -231,9 +231,8 @@ def run_customized_training_loop(
else: else:
train_summary_writer = None train_summary_writer = None
# De-dupes variables due to keras tracking issues. # Collects training variables.
training_vars = list({id(v): v for v in model.trainable_variables training_vars = model.trainable_variables
}.values())
def _replicated_step(inputs): def _replicated_step(inputs):
"""Replicated training step.""" """Replicated training step."""
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment