Commit 20ed8cf0 authored by Reed Wanderman-Milne's avatar Reed Wanderman-Milne Committed by A. Unique TensorFlower
Browse files

Remove dynamic_loss_scale argument to define_performance.

All models which support loss scaling support dynamic loss scaling, so the argument has no purpose. It used to be that some models scaled the loss manually instead of using a LossScaleOptimizer, and so did not support dynamic loss scaling.

PiperOrigin-RevId: 367719521
parent 4e47fe2b
......@@ -267,7 +267,7 @@ def run(flags_obj):
def define_cifar_flags():
common.define_keras_flags(dynamic_loss_scale=False)
common.define_keras_flags()
flags_core.set_defaults(data_dir='/tmp/cifar10_data/cifar-10-batches-bin',
model_dir='/tmp/cifar10_model',
......
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