nianticlabs / simplerecon

[ECCV 2022] SimpleRecon: 3D Reconstruction Without 3D Convolutions
Other
1.31k stars 121 forks source link

Hosting models on the Hugging Face Hub #5

Open NimaBoscarino opened 2 years ago

NimaBoscarino commented 2 years ago

Hello! Thank you so much for open-sourcing this work, this is amazing 🙏

I was wondering if you'd be interested in mirroring your pretrained models over on the Hugging Face model hub, and possibly even host a demo on Hugging Face Spaces. Putting models on the Hub helps our community discover them, and showcasing them with demos has been a really great way to reach out to the broader tech world! Some demos have even gone pretty viral, like Stable Diffusion.

We've got guides on how to upload models and on creating spaces, but I'm more than happy to help you out with this stuff If you're interested! We could even see if it's possible to secure a GPU for your demo.

Let me know what you think, and thanks again for open-sourcing this!

mohammed-amr commented 2 years ago

Hello hello! Welcome and thanks! glad people are finding it useful.

Thanks for reaching out and suggesting this! I definitely see the appeal and the benefit this could have. I'll have to discuss this with my team and see if it's something we can support alongside the Github repo.

I'm not totally familiar with Spaces's ability. Can users upload a zip file with data that the model is allowed access to? I can see a version of this where users upload a raw data from Scanniverse that can help in this.

Thanks!

NimaBoscarino commented 2 years ago

Awesome! Yes, a user could definitely upload a file to a Space using Gradio's File Input component 😊

Let us know if we can offer any guidance!