metadata
dataset_info:
features:
- name: index
dtype: int64
- name: report_id
dtype: int64
- name: subject_id
dtype: int64
- name: raw_text
dtype: string
- name: raw_len
dtype: int64
- name: history
dtype: string
- name: comparison
dtype: string
- name: procedure
dtype: string
- name: findings
dtype: string
- name: impression
dtype: string
- name: source
dtype: string
- name: has_impression
dtype: bool
- name: has_findings
dtype: bool
- name: has_history
dtype: bool
- name: has_comparison
dtype: bool
- name: has_procedure
dtype: bool
- name: num_sections_found
dtype: int64
- name: missing_sections
dtype: string
- name: extraction_notes
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 754695629
num_examples: 233281
- name: validation
num_bytes: 93695014
num_examples: 29160
- name: test
num_bytes: 94162497
num_examples: 29161
download_size: 410456466
dataset_size: 942553140
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*