metadata
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 115290368
num_examples: 100
download_size: 115287242
dataset_size: 115290368
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
This dataset contains 50 images collected from open sources on the Internet for the query "iris" and 50 synthetic images created by ChatGPT-4o. All images are annotated by ChatGPT-4o.
(Beautiful eyes - synthetic)