Datasets:

StellarMilk
/

squad_newsqa

language:
  - en
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/squad_newsqa_train.parquet
      - split: validation
        path: data/squad_newsqa_valid.parquet
      - split: test
        path: data/squad_newsqa_test.parquet
size_categories:
  - 10K<n<100K

squad_newsqa

Concatenating StellarMilk/newsqa dataset with lmqg/qag_squad for asahi417/lm-question-generation