metadata
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: 109468910.646
num_examples: 3102
- name: validation
num_bytes: 19169372
num_examples: 546
- name: test
num_bytes: 30201592
num_examples: 886
download_size: 158656963
dataset_size: 158839874.646
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*