metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: audio_path
dtype: string
- name: sampling_rate
dtype: int32
- name: beat
sequence: float32
- name: downbeat
sequence: float32
- name: meter
dtype: string
- name: tempo
dtype: float32
- name: genre
dtype: string
- name: duration
dtype: float32
- name: split
dtype: string
splits:
- name: train
num_bytes: 186520
num_examples: 442
- name: validation
num_bytes: 86120
num_examples: 197
- name: test
num_bytes: 120230
num_examples: 290
download_size: 321230
dataset_size: 392870
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*