README.md

---
dataset_info:
  features:
  - name: molecule
    dtype: string
  - name: caption
    dtype: string
  splits:
  - name: train
    num_bytes: 276260470
    num_examples: 802800
  - name: split_train
    num_bytes: 219611475
    num_examples: 634320
  - name: split_valid
    num_bytes: 56648995
    num_examples: 168480
  download_size: 78056020
  dataset_size: 552520940
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: split_train
    path: data/split_train-*
  - split: split_valid
    path: data/split_valid-*
---