metadata
dataset_info:
features:
- name: filename
dtype: string
- name: timestamp
dtype: timestamp[ns]
- name: image
dtype: image
- name: score
dtype: float64
- name: image_index
dtype: int64
- name: walk_index
dtype: int64
- name: walk_name
dtype: string
- name: D voted
dtype: bool
- name: A voted
dtype: bool
- name: G voted
dtype: bool
- name: A_score
dtype: 'null'
- name: D_score
dtype: float64
- name: G_score
dtype: float64
splits:
- name: train
num_bytes: 94670080
num_examples: 606
download_size: 94651500
dataset_size: 94670080
configs:
- config_name: default
data_files:
- split: train
path: data/train-*