README.md

---
dataset_info:
  features:
  - name: track_id
    dtype: string
  - name: tag_list
    sequence: string
  - name: pseudo_caption
    dtype: string
  - name: title
    dtype: string
  - name: artist_name
    dtype: string
  - name: release
    dtype: string
  - name: year
    dtype: string
  - name: path
    dtype: string
  splits:
  - name: train
    num_bytes: 19274195
    num_examples: 81438
  - name: valid
    num_bytes: 2582829
    num_examples: 10801
  - name: test
    num_bytes: 2629913
    num_examples: 11064
  download_size: 11675639
  dataset_size: 24486937
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
  - split: test
    path: data/test-*
---