metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype: int64
- name: speaker
dtype: int64
- name: full_path
dtype: string
- name: arousal
dtype: float64
- name: valence
dtype: float64
splits:
- name: train
num_bytes: 190239959.23361388
num_examples: 17944
- name: validation
num_bytes: 47559846.25640346
num_examples: 4486
- name: test
num_bytes: 65505776.83698266
num_examples: 6177
download_size: 302504742
dataset_size: 303305582.327
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*