README.md

metadata

dataset_info:
  features:
    - name: Canonicalized SMILES
      dtype: string
    - name: Canonicalized Flavor
      dtype: string
    - name: Original Labels
      dtype: string
    - name: Source
      dtype: string
  splits:
    - name: train
      num_bytes: 1970859
      num_examples: 22038
    - name: test
      num_bytes: 420863
      num_examples: 4722
    - name: validation
      num_bytes: 419548
      num_examples: 4723
  download_size: 778401
  dataset_size: 2811270
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
      - split: validation
        path: data/validation-*