---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: arousal
    dtype: float64
  - name: valence
    dtype: float64
  - name: label
    dtype: int64
  - name: arousal_norm
    dtype: float64
  - name: valence_norm
    dtype: float64
  - name: speakerID
    dtype: float64
  splits:
  - name: train
    num_bytes: 3946919880.798199
    num_examples: 83919
  - name: test
    num_bytes: 840089584.63
    num_examples: 17789
  - name: validation
    num_bytes: 788975580.31
    num_examples: 16786
  download_size: 5571456539
  dataset_size: 5575985045.738199
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: validation
    path: data/validation-*
---