metadata
dataset_info:
features:
- name: SELFIES
dtype: string
- name: molecule_smiles
dtype: string
- name: input_ids
sequence: int32
- name: token_type_ids
sequence: int8
- name: special_tokens_mask
sequence: int8
- name: attention_mask
sequence: int8
- name: length
dtype: int64
splits:
- name: train
num_bytes: 49738795393.73214
num_examples: 89364268
- name: test
num_bytes: 5526533316.267859
num_examples: 9929364
download_size: 16925243419
dataset_size: 55265328710
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*