metadata
dataset_info:
features:
- name: text
dtype: string
- name: genre
dtype: string
- name: cp
dtype: string
- name: key
dtype: string
- name: bpm
dtype: string
- name: time_signature
dtype: string
- name: fname
dtype: string
splits:
- name: train
num_bytes: 345194639
num_examples: 9024
download_size: 55710653
dataset_size: 345194639
configs:
- config_name: default
data_files:
- split: train
path: data/train-*