---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: label
    dtype: int64
  - name: valence
    dtype: float64
  - name: arousal
    dtype: float64
  - name: domination
    dtype: float64
  - name: arousal_norm
    dtype: float64
  - name: valence_norm
    dtype: float64
  - name: speakerID
    dtype: int64
  splits:
  - name: train
    num_bytes: 250962217.02250355
    num_examples: 2387
  - name: validation
    num_bytes: 56302333.18681318
    num_examples: 540
  - name: test
    num_bytes: 90762330.21331829
    num_examples: 879
  download_size: 399855622
  dataset_size: 398026880.422635
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---