metadata
dataset_info:
features:
- name: instruction
dtype: string
- name: task_output
dtype: string
- name: transcription
dtype: string
- name: description
dtype: string
- name: medical_specialty
dtype: string
- name: sample_name
dtype: string
- name: keywords
dtype: string
- name: derived_keywords
dtype: string
- name: transcription_length
dtype: int64
- name: normalized_length
dtype: float64
- name: complexity_score
dtype: float64
- name: text
dtype: string
splits:
- name: train
num_bytes: 21026370
num_examples: 3000
download_size: 9959099
dataset_size: 21026370
configs:
- config_name: default
data_files:
- split: train
path: data/train-*