---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: smiles
    dtype: string
  - name: caption
    dtype: string
  - name: selfies
    dtype: string
  - name: properties
    sequence: string
  splits:
  - name: train
    num_bytes: 2281977596.268382
    num_examples: 126863
  - name: validation
    num_bytes: 608696992.0944623
    num_examples: 33695
  download_size: 2756329992
  dataset_size: 2890674588.3628445
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---