Datasets:

Modalities:
Text
Formats:
parquet
Languages:
English
ArXiv:
Libraries:
Datasets
pandas
License:
bruel's picture
Upload dataset
c9e0d9f verified
|
raw
history blame
574 Bytes
metadata
dataset_info:
  features:
    - name: input
      dtype: string
    - name: output
      sequence: string
    - name: id
      dtype: string
  splits:
    - name: train
      num_bytes: 2226504.859142607
      num_examples: 2743
    - name: test
      num_bytes: 278414.5704286964
      num_examples: 343
    - name: valid
      num_bytes: 278414.5704286964
      num_examples: 343
  download_size: 543268
  dataset_size: 2783334
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
      - split: valid
        path: data/valid-*