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

Fix wrong argument bug in PPOv2Trainer #1660

Closed
wants to merge 3 commits into from
Closed

Fix wrong argument bug in PPOv2Trainer #1660

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
e2650e6
fix inheritance order in PPOv2Config
Nicolinho May 23, 2024
f50753c
fix inheritance order in rloo_config
Nicolinho May 23, 2024
46c5dbb
fix wrong argument bug in PPOv2Trainer
Nicolinho May 23, 2024
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
2 changes: 1 addition & 1 deletion trl/trainer/ppov2_config.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@


@dataclass
class PPOv2Config(TrainingArguments, OnpolicyRuntimeConfig):
class PPOv2Config(OnpolicyRuntimeConfig, TrainingArguments):
# common config
exp_name: str = os.path.basename(__file__)[: -len(".py")]
"""the name of this experiment"""
Expand Down
4 changes: 2 additions & 2 deletions trl/trainer/ppov2_trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -285,7 +285,7 @@ def repeat_generator():
query_response, logits = generate(
unwrapped_model.policy,
query,
tokenizer,
tokenizer.pad_token_id,
generation_config,
)
response = query_response[:, context_length:]
Expand Down Expand Up @@ -407,7 +407,7 @@ def repeat_generator():
mb_query_responses = query_responses[micro_batch_inds]
mb_logprobs = logprobs[micro_batch_inds]

output, vpred_temp = forward(model, mb_query_responses, tokenizer)
output, vpred_temp = forward(model, mb_query_responses, tokenizer.pad_token_id)
logits = output.logits[:, context_length - 1 : -1]
logits /= args.temperature + 1e-7
new_all_logprobs = F.log_softmax(logits, dim=-1)
Expand Down
2 changes: 1 addition & 1 deletion trl/trainer/rloo_config.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@


@dataclass
class RLOOConfig(TrainingArguments, OnpolicyRuntimeConfig):
class RLOOConfig(OnpolicyRuntimeConfig, TrainingArguments):
# common config
exp_name: str = os.path.basename(__file__)[: -len(".py")]
"""the name of this experiment"""
Expand Down