Datasets, models and summary added to text-to-image task #378
Conversation
There was a problem hiding this comment.
Thanks a lot for contributions!!! ❤️ I left two comments, otherwise looks very good to me 🙂
The pages serve as the first stop people who would like to know about a task and maybe build a quick PoC for their business case on it so it's important to stay a bit more general on the models and the datasets side 🙂
There was a problem hiding this comment.
Overall looks good to me, @osanseviero can you quickly check? I don't think we have a canonical dataset for the task BTW (not very sure)
I found two image-to-text ones which can be used for the task imo, WDYT:
https://huggingface.co/datasets/red_caps
https://huggingface.co/datasets/conceptual_captions
I'll be happy to merge after the key-space and indent errors are fixed 🙂
There was a problem hiding this comment.
Great contribution, thanks! 🔥
As @merveenoyan mentioned, a couple of datasets could be a good fit here. Apart from the 2 already great ones Merve shared, the ones from LAION are quite used right now https://huggingface.co/laion (e.g. https://huggingface.co/datasets/laion/laion2B-en-aesthetic)
Co-authored-by: Omar Sanseviero <[email protected]>
I have fixed the indent and key-value space issues but i couldnt find the eslint config file. Could you give the configuration file you have used for future uses ? |
|
I will format it myself, thanks a lot for the contribution! What is your hugging face Hub handle so I can add your name to the doc later? 🙂 @oguzakif |
|
Did you mean my hugging face username ? https://huggingface.co/oguzakif
|
3 participants
Summary