Skip to content

Conversation

@yhrkw
Copy link
Contributor

@yhrkw yhrkw commented Aug 26, 2025

In several models, they fail during generation and raise RuntimeError. This appears to occur with both pre-trained models and API-based models.

We will include generation success/failure status to trial_*.json while excluding it from score calculations.

json.dump(data, f, indent=2, ensure_ascii=False)

result["num_trials"] = min([len(x) for x in data.values()])
result["num_trials"] = max([len(x) for x in data.values()])
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Putting num_trials in config.json seems appropriate.

@yhrkw yhrkw force-pushed the unraise-not-answered branch from 7ac75fd to 6fb652e Compare September 4, 2025 08:16
@yhrkw yhrkw requested a review from imos September 4, 2025 08:18
@imos imos merged commit c3bd1cb into pfnet-research:main Sep 4, 2025
1 check passed
@yhrkw yhrkw deleted the unraise-not-answered branch September 17, 2025 00:10
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants