Skip to content

Show intermediate reward in vf-eval #693

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
willccbb merged 1 commit into from
Jan 6, 2026
Merged

Show intermediate reward in vf-eval #693

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
11 changes: 11 additions & 0 deletions verifiers/envs/environment.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -867,8 +867,10 @@ async def generate(
pbar = tqdm(
total=len(group_list),
desc=f"Processing {len(group_list)} groups ({len(inputs_list)} total rollouts)",
postfix=dict(reward="?"),
)

reward_sum, reward_count = 0, 0
groups_completed = 0
all_states: list[State] = []
try:
Expand All @@ -877,8 +879,17 @@ async def generate(
all_states.extend(group_states)
groups_completed += 1

# track reward for rolling average
for s in group_states:
r = s.get("reward")
if r is not None:
reward_sum += r
reward_count += 1

if pbar is not None:
pbar.update(1)
if reward_count > 0:
pbar.set_postfix(reward=f"{reward_sum / reward_count:.3f}")

# save intermediate results
if (
Expand Down
Loading