metadata
dataset_info:
features:
- name: subject_id
dtype: int64
- name: hadm_id
dtype: int64
- name: text
dtype: string
- name: icd_code
sequence: string
- name: cleaned_text
dtype: string
- name: code_rare
sequence: string
- name: code_commons
sequence: string
- name: icd_descriptions
sequence: string
splits:
- name: train
num_bytes: 2914958674
num_examples: 122288
download_size: 1478570623
dataset_size: 2914958674
configs:
- config_name: default
data_files:
- split: train
path: data/train-*