metadata
dataset_info:
features:
- name: subject_id
dtype: int64
- name: note_id
dtype: string
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 3574282714
num_examples: 134100
- name: validation
num_bytes: 159548049
num_examples: 5986
- name: test
num_bytes: 300037102
num_examples: 11257
download_size: 614312378
dataset_size: 4033867865
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*