Skip to content

cp: fix: Fix the logger error in non-colocated sync-grpo code path (1355) into r0.4.0 #1370

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

Merged
terrykong merged 1 commit into from
Oct 16, 2025
Merged

cp: fix: Fix the logger error in non-colocated sync-grpo code path (1355) into r0.4.0 #1370

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
3 changes: 0 additions & 3 deletions nemo_rl/algorithms/grpo.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1779,9 +1779,6 @@ def async_grpo_train(
train_results, metrics, timing_metrics, master_config
)

if "per_worker_token_counts" in metrics:
del metrics["per_worker_token_counts"]

logger.log_metrics(performance_metrics, step + 1, prefix="performance")
logger.log_metrics(metrics, step + 1, prefix="train")
logger.log_metrics(timing_metrics, step + 1, prefix="timing/train")
Expand Down
9 changes: 9 additions & 0 deletions nemo_rl/algorithms/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -553,6 +553,15 @@ def visualize_per_worker_load(per_worker_token_counts: dict[int, int]) -> float:
total_tflops / theoretical_tflops
)

# =====================================================
# Clean up metrics
# =====================================================

# Clean up metrics to avoid wandb logging errors
# Dict structures cannot be logged to wandb
if "per_worker_token_counts" in metrics:
del metrics["per_worker_token_counts"]

# =====================================================
# Logging
# =====================================================
Expand Down
Loading