Datasets:

tattrongvu
/

sharegpt4v_vqa_f200k

dataset_info:
  features:
    - name: image
      dtype: image
    - name: image_path
      dtype: string
    - name: query
      dtype: string
  splits:
    - name: train
      num_bytes: 197999024151
      num_examples: 200000
  download_size: 197890620578
  dataset_size: 197999024151
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: mit
task_categories:
  - image-to-text
language:
  - en
size_categories:
  - 100K<n<1M