---
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
  - name: utterance_id
    dtype: string
  - name: transcript
    dtype: string
  splits:
  - name: train
    num_bytes: 324012945.276
    num_examples: 3102
  - name: validation
    num_bytes: 56973934.0
    num_examples: 546
  - name: test
    num_bytes: 91558942.0
    num_examples: 886
  download_size: 474597552
  dataset_size: 472545821.276
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---