Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

fix repetition penalty error in modeling_utils.py #2303

Merged
merged 1 commit into from
Dec 25, 2019
Merged

fix repetition penalty error in modeling_utils.py #2303

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 10 additions & 2 deletions src/transformers/modeling_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -728,7 +728,11 @@ def _generate_no_beam_search(
if repetition_penalty != 1.0:
for i in range(batch_size):
for previous_tokens in set(input_ids[i].tolist()):
next_token_logits[i, previous_tokens] /= repetition_penalty
# if score < 0 then repetition penalty has to multiplied to reduce the previous token probability
if next_token_logits[i, previous_tokens] < 0:
next_token_logits[i, previous_tokens] *= repetition_penalty
else:
next_token_logits[i, previous_tokens] /= repetition_penalty

if do_sample:
# Temperature (higher temperature => more likely to sample low probability tokens)
Expand Down Expand Up @@ -807,7 +811,11 @@ def _generate_beam_search(
if repetition_penalty != 1.0:
for i in range(batch_size * num_beams):
for previous_tokens in set(input_ids[i].tolist()):
scores[i, previous_tokens] /= repetition_penalty
# if score < 0 then repetition penalty has to multiplied to reduce the previous token probability
if scores[i, previous_tokens] < 0:
scores[i, previous_tokens] *= repetition_penalty
else:
scores[i, previous_tokens] /= repetition_penalty

if do_sample:
# Temperature (higher temperature => more likely to sample low probability tokens)
Expand Down