Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
ModelZoo
ResNet50_tensorflow
Commits
e8140fa9
Commit
e8140fa9
authored
Feb 12, 2020
by
Chen Chen
Committed by
A. Unique TensorFlower
Feb 12, 2020
Browse files
Remove output_activation because it is not used anywhere.
PiperOrigin-RevId: 294813724
parent
35aa1f31
Changes
1
Show whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
3 additions
and
5 deletions
+3
-5
official/nlp/modeling/networks/bert_pretrainer.py
official/nlp/modeling/networks/bert_pretrainer.py
+3
-5
No files found.
official/nlp/modeling/networks/bert_pretrainer.py
View file @
e8140fa9
...
...
@@ -33,9 +33,9 @@ class BertPretrainer(tf.keras.Model):
encoder as described in "BERT: Pre-training of Deep Bidirectional Transformers
for Language Understanding" (https://arxiv.org/abs/1810.04805).
The Bert
T
rainer allows a user to pass in a transformer stack, and
instantiates
the masked language model and classification networks that are
used to create
the training objectives.
The Bert
Pret
rainer allows a user to pass in a transformer stack, and
instantiates
the masked language model and classification networks that are
used to create
the training objectives.
Attributes:
network: A transformer network. This network should output a sequence output
...
...
@@ -56,7 +56,6 @@ class BertPretrainer(tf.keras.Model):
num_classes
,
num_token_predictions
,
activation
=
None
,
output_activation
=
None
,
initializer
=
'glorot_uniform'
,
output
=
'logits'
,
**
kwargs
):
...
...
@@ -66,7 +65,6 @@ class BertPretrainer(tf.keras.Model):
'num_classes'
:
num_classes
,
'num_token_predictions'
:
num_token_predictions
,
'activation'
:
activation
,
'output_activation'
:
output_activation
,
'initializer'
:
initializer
,
'output'
:
output
,
}
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment