hi-en-dataset / README.md
geekyrakshit's picture
Upload dataset
a949850 verified
|
raw
history blame
384 Bytes
metadata
dataset_info:
  features:
    - name: hi
      dtype: string
    - name: en
      dtype: string
    - name: en_embedding
      sequence:
        sequence:
          sequence: float32
  splits:
    - name: train
      num_bytes: 31578005
      num_examples: 100
  download_size: 32159877
  dataset_size: 31578005
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*