NousResearch/atropos
1
0
Fork 0
mirror of https://github.com/NousResearch/atropos.git synced 2026-04-19 12:57:58 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

more linter nonsense

Browse source
This commit is contained in:
teknium 2025-12-24 11:04:33 +00:00
parent f18d46549d
commit abdda3978a
29 changed files with 113 additions and 151 deletions
Download patch file Download diff file Expand all files Collapse all files

4
environments/eval_environments/hle_eval.py
View file

@ -152,7 +152,7 @@ class HLEEvalEnv(BaseEnv):
if not self._dataset_loaded:
await self._load_dataset()
print(f"\nHLE Evaluation Setup (Generative Mode):")
print("\nHLE Evaluation Setup (Generative Mode):")
print(f" Dataset: {self.config.dataset_name}")
print(f" Evaluation split: {self.config.eval_split}")
print(f" Thinking mode: {self.config.thinking_mode}")
@ -525,7 +525,7 @@ class HLEEvalEnv(BaseEnv):
if self.config.thinking_mode:
print(f" Thinking Utilization: {has_thinking / total:.2%}")
if category_metrics:
print(f"\n Per-Category Breakdown:")
print("\n Per-Category Breakdown:")
for cat, data in sorted(
category_metrics.items(), key=lambda x: -x[1]["accuracy"]
):