metadata
dataset_info:
features:
- name: user
dtype: string
- name: image_id
dtype: int64
- name: original_image_path
dtype: string
- name: original_image
dtype: image
- name: langsam_res
dtype: string
- name: caption
dtype: string
- name: brand
dtype: string
- name: quality
dtype: string
- name: lifestyle
dtype: bool
- name: product
dtype: bool
- name: text
dtype: bool
- name: frr_image
dtype: image
- name: masks
dtype: string
splits:
- name: train
num_bytes: 7383918
num_examples: 2
download_size: 515422
dataset_size: 7383918
configs:
- config_name: default
data_files:
- split: train
path: data/train-*