license: wtfpl
task_categories:
- image-classification
- image-segmentation
- image-to-text
- text-to-image
- image-feature-extraction
tags:
- not-for-all-audiences
pretty_name: The Unlaion Temple
size_categories:
- 1B<n<10B
All files uploaded. Enjoy!
Dataset Card for The Unlaion Temple
Dataset Details
Dataset Description
Laion-5B is still not public, so we decided to create our own dataset.
The Unlaion Temple is a raw dataset of CommonCrawl images (Estimated to be a total of 2 Billion urls). We haven't verified whether the links in this dataset are functional.
You are responsible for handling the data.
We've made some improvements to the dataset based on user feedback:
- All images have been captured, including alt text.
- Figure images and captions have been added. We have not checked if this works.
If you ever feel uncomfortable while downloading the dataset, pause the download and take a break.
Go outside and get some fresh air by opening the front door and taking a walk.
Uses
We strongly advise against using this dataset.
Copyright
If you encounter copyright issues with this dataset, please refrain from blaming the dataset creators.
Instead, contact the website hosting the file and request its removal. This approach is more effective and should be followed.
Special Notice
In light of recent AI Developments, we define the additional clauses:
- This applies only to the following corporations, organizations and thier related subsidaries as listed:
- Alphabet, Amazon, Apple, Meta, Microsoft, Baidu, Alibaba, Tencent, Huawei, Xiaomi, OpenAI, Stability AI, RunwayML, Anthropic, Midjourney
- Upon utilizing this dataset, you hereby affirm and declare that you:
- Do not represent nor are affiliated with aforementioned corporations and organizations.
- Assert that your use of this data is independent of any association with these entities.
- To the extent of your understanding, belief, and knowledge, you guarantee that this dataset will not find its way into the possession of the aforementioned corporations and organizations through your actions.
- By proceeding, you acknowledge and adhere to these terms, ensuring that your utilization of the data remains separate from any influence or connection to the aforementioned corporations and organizations.
- In demonstrating your commitment to open-source principles, you pledge to publicly disclose all resulting datasets and scripts within 62 days (1,488 hours) of finalizing your model or pre-print, which ever comes first.
- Any members in the Open Model Initiative are not subjected to the following restrictions.