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
chenpangpang
transformers
Commits
6e4c8f68
Unverified
Commit
6e4c8f68
authored
Oct 11, 2021
by
Jungwoo Park
Committed by
GitHub
Oct 11, 2021
Browse files
change to apply `pad_to_multiple_of` to labels (#13949)
parent
dca67968
Changes
1
Show whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
7 additions
and
0 deletions
+7
-0
src/transformers/data/data_collator.py
src/transformers/data/data_collator.py
+7
-0
No files found.
src/transformers/data/data_collator.py
View file @
6e4c8f68
...
@@ -534,6 +534,13 @@ class DataCollatorForSeq2Seq:
...
@@ -534,6 +534,13 @@ class DataCollatorForSeq2Seq:
# same length to return tensors.
# same length to return tensors.
if
labels
is
not
None
:
if
labels
is
not
None
:
max_label_length
=
max
(
len
(
l
)
for
l
in
labels
)
max_label_length
=
max
(
len
(
l
)
for
l
in
labels
)
if
self
.
pad_to_multiple_of
is
not
None
:
max_label_length
=
(
(
max_label_length
+
self
.
pad_to_multiple_of
-
1
)
//
self
.
pad_to_multiple_of
*
self
.
pad_to_multiple_of
)
padding_side
=
self
.
tokenizer
.
padding_side
padding_side
=
self
.
tokenizer
.
padding_side
for
feature
in
features
:
for
feature
in
features
:
remainder
=
[
self
.
label_pad_token_id
]
*
(
max_label_length
-
len
(
feature
[
"labels"
]))
remainder
=
[
self
.
label_pad_token_id
]
*
(
max_label_length
-
len
(
feature
[
"labels"
]))
...
...
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