Skip to content

fix: load GAIA from Parquet dataset - #273

Draft
ruirui6946 wants to merge 1 commit into
TencentCloudADP:mainfrom
ruirui6946:agent/fix-gaia-parquet-loader
Draft

fix: load GAIA from Parquet dataset#273
ruirui6946 wants to merge 1 commit into
TencentCloudADP:mainfrom
ruirui6946:agent/fix-gaia-parquet-loader

Conversation

@ruirui6946

Copy link
Copy Markdown

Summary

  • load the current GAIA Parquet snapshot through its dataset configuration instead of the removed GAIA.py loader
  • keep the dataset README so Hugging Face can discover the available configurations
  • make the import script safe to import and straightforward to unit test
  • document GAIA access, import, and evaluation commands

Why

The upstream GAIA dataset migrated to Parquet and no longer ships GAIA.py. The existing importer therefore fails before any examples are processed.

Closes #245

Validation

  • ruff check scripts/data/process_gaia.py tests/scripts/test_process_gaia.py
  • ruff format --check scripts/data/process_gaia.py tests/scripts/test_process_gaia.py
  • pytest -p no:cacheprovider tests/scripts/test_process_gaia.py -q
  • mkdocs build --strict

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.

gaia-benchmark 导入数据集报错

1 participant