---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: speaker
    dtype: string
  - name: speakerID
    dtype: int64
  - name: label
    dtype: int64
  - name: dataset
    dtype: string
  - name: valence
    dtype: float64
  - name: arousal
    dtype: float64
  - name: domination
    dtype: float64
  - name: arousal_norm
    dtype: float64
  - name: valence_norm
    dtype: float64
  - name: file_paths
    dtype: string
  splits:
  - name: train
    num_bytes: 343541098.0
    num_examples: 3599
  download_size: 343052485
  dataset_size: 343541098.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---