dataset_info:
features:
- name: are_different
dtype: bool
- name: best_image_uid
dtype: string
- name: caption
dtype: string
- name: created_at
dtype: timestamp[ns]
- name: has_label
dtype: bool
- name: image_0_uid
dtype: string
- name: image_0_url
dtype: string
- name: image_1_uid
dtype: string
- name: image_1_url
dtype: string
- name: jpg_0
dtype: binary
- name: jpg_1
dtype: binary
- name: label_0
dtype: float64
- name: label_1
dtype: float64
- name: model_0
dtype: string
- name: model_1
dtype: string
- name: ranking_id
dtype: int64
- name: user_id
dtype: int64
- name: num_example_per_prompt
dtype: int64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 193273338802
num_examples: 583747
- name: validation
num_bytes: 5638295249
num_examples: 17439
- name: test
num_bytes: 4621428929
num_examples: 14073
- name: validation_unique
num_bytes: 178723392
num_examples: 500
- name: test_unique
num_bytes: 178099641
num_examples: 500
download_size: 202289408791
dataset_size: 203889886013
Dataset Card for Pick-a-Pic (v1)
Dataset Description
- Homepage: The web app can be found at pickapic.io
- Repository: The repository of PickScore
- Paper: Pick-a-Pic: An Open Dataset of User Preferences for Text-to-Image Generation.
- **Leaderboard: TODO **
- **Point of Contact: TODO **
Dataset Summary
The Pick-a-Pic dataset was collected with the Pick-a-Pic web app and contains over half-a-million examples of human preferences over model-generated images. This dataset with URLs instead of the actual images (which makes it much smaller in size) can be found here.
See the corresponding paper Pick-a-Pic: An Open Dataset of User Preferences for Text-to-Image Generation for more details.
If you want to download this dataset with URLs instead of images to save space, please see this version of the dataset.
Supported Tasks and Leaderboards
Task: Select preferred image in test-set.
| Models | Test-Set Accuracy (%) |
|---|---|
| PickScore | 70.2% |
| Human Expert Baseline | 68.0% |
| HPS | 66.7% |
| ImageReward | 61.1% |
| CLIP-H | 60.8% |
| Aesthetics | 56.8% |
Data Splits
The dataset has three main splits: train, validation, validation_unique (with one example per prompt), test, and test_unique.
Citation Information
If you find this work useful, please cite:
@inproceedings{Kirstain2023PickaPicAO,
title={Pick-a-Pic: An Open Dataset of User Preferences for Text-to-Image Generation},
author={Yuval Kirstain and Adam Polyak and Uriel Singer and Shahbuland Matiana and Joe Penna and Omer Levy},
year={2023}
}
LICENSE
MIT License
Copyright (c) 2021
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.