metadata
dataset_info:
features:
- name: participant_id
dtype: string
- name: pair_id
dtype: int64
- name: company
dtype: string
- name: screen_type
dtype: string
- name: target_description
dtype: string
- name: chosen_kind
dtype: string
- name: chosen_image
dtype: image
- name: chosen_svg
dtype: string
- name: rejected_kind
dtype: string
- name: rejected_image
dtype: image
- name: rejected_svg
dtype: string
- name: reference_images
list: image
- name: reference_svgs
list: string
- name: created_at
dtype: string
splits:
- name: train
num_bytes: 870867593
num_examples: 70
download_size: 869079976
dataset_size: 870867593
configs:
- config_name: default
data_files:
- split: train
path: data/train-*