Skip to content

fix: validate ngram_size in get_repetition_penalty_reward #7129

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

Closed
CJstate wants to merge 1 commit into from
+7 −0
Closed

fix: validate ngram_size in get_repetition_penalty_reward #7129

Show file tree
Hide file tree
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
5 changes: 5 additions & 0 deletions tests/test_rewards.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -161,6 +161,11 @@ def test_positive_max_penalty_raises(self):
with pytest.raises(ValueError):
get_repetition_penalty_reward(ngram_size=2, max_penalty=0.5)

@pytest.mark.parametrize("ngram_size", [0, -1])
def test_non_positive_ngram_size_raises(self, ngram_size):
with pytest.raises(ValueError):
get_repetition_penalty_reward(ngram_size=ngram_size, max_penalty=-1.0)

def test_extra_kwargs_are_ignored(self):
"""Trainers pass prompts/completions/etc. as kwargs; the reward must accept and ignore them."""
reward_fn = get_repetition_penalty_reward(ngram_size=2, max_penalty=-1.0)
Expand Down
2 changes: 2 additions & 0 deletions trl/rewards/other_rewards.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,6 +56,8 @@ def get_repetition_penalty_reward(ngram_size: int = 3, max_penalty: float = -1.0
"""
if max_penalty > 0:
raise ValueError(f"max_penalty {max_penalty} should not be positive")
if ngram_size <= 0:
raise ValueError(f"ngram_size {ngram_size} should be greater than 0")
return _RepetitionPenalty(ngram_size, max_penalty)


Expand Down
Loading