README.md

metadata

configs:
  - config_name: default
    data_files:
      - split: train
        path: data.csv
      - split: test
        path: holdout.csv
size_categories:
  - n<1K

[doc] manual configuration 1

This dataset contains two csv files at the root, and a YAML field configs that specifies the data files and splits.