Skip to content

GitLab

Commit c25d7711 authored by Hongkun Yu's avatar Hongkun Yu Committed by A. Unique TensorFlower
Browse files

tf.keras.backend.sparse_categorical_crossentropy -> tf.keras.losses.sparse_categorical_crossentropy 

PiperOrigin-RevId: 296755118
parent 6f0e3a0b
Hide whitespace changes
Inline Side-by-side
Showing with 2 additions and 2 deletions
official/nlp/bert/run_squad.py
View file @ c25d7711
...@@ -109,9 +109,9 @@ def squad_loss_fn(start_positions, ...@@ -109,9 +109,9 @@ def squad_loss_fn(start_positions,
end_logits, end_logits,
loss_factor=1.0): loss_factor=1.0):
"""Returns sparse categorical crossentropy for start/end logits.""" """Returns sparse categorical crossentropy for start/end logits."""
start_loss = tf.keras.backend.sparse_categorical_crossentropy( start_loss = tf.keras.losses.sparse_categorical_crossentropy(
start_positions, start_logits, from_logits=True) start_positions, start_logits, from_logits=True)
end_loss = tf.keras.backend.sparse_categorical_crossentropy( end_loss = tf.keras.losses.sparse_categorical_crossentropy(
end_positions, end_logits, from_logits=True) end_positions, end_logits, from_logits=True)
total_loss = (tf.reduce_mean(start_loss) + tf.reduce_mean(end_loss)) / 2 total_loss = (tf.reduce_mean(start_loss) + tf.reduce_mean(end_loss)) / 2
......
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