metadata
dataset_info:
features:
- name: post_id
dtype: string
- name: image
dtype: image
- name: instruction
dtype: string
- name: simplified_instruction
dtype: string
- name: creativity_index
dtype:
class_label:
names:
'0': Low
'1': Medium
'2': High
- name: object_list
sequence: string
splits:
- name: train
num_bytes: 161339488
num_examples: 300
download_size: 160496682
dataset_size: 161339488
configs:
- config_name: default
data_files:
- split: train
path: data/train-*