Instructions to use facebook/EUPE-ViT-T with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- EUPE
How to use facebook/EUPE-ViT-T with EUPE:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Add transformers-format weights (safetensors) + config
This PR adds a transformers-compatible version of the checkpoint: safetensors weights, config.json, image-processor config, and a model card with library_name: transformers. The architecture reuses DINOv3 (DINOv3ViTModel / DINOv3ConvNextModel), so the model loads with AutoModel.from_pretrained(...). Weights were converted from the original .pt by renaming the research key names to the transformers module names; outputs match the original forward_features numerically. Opening as a PR for the authors to review.
FYI @zcckernel π€ I'll open also PRs on the 5 other repos. This will allow direct use of EUPE in transformers once https://github.com/huggingface/transformers/pull/46691 is merged. Thank you!
Edit: We don't even need the PR mentioned to be merged, it's just to trace how conversion to .safetensors format was done