---
dataset_info:
  features:
  - name: item_id
    dtype: string
  - name: start
    dtype: string
  - name: freq
    dtype: string
  - name: target
    sequence: float32
  splits:
  - name: train
    num_bytes: 3726883897.65553
    num_examples: 29478
  download_size: 1807987455
  dataset_size: 3726883897.65553
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---