Instructions to use RicePasteM/MeowID-Base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- TensorRT
How to use RicePasteM/MeowID-Base with TensorRT:
# 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
Missing cat_recognition.data module needed for training
Hi, and thanks for releasing MeowID. We run a community cat program in Brooklyn (Flatbush Cats) and we’re using MeowID to match photos of cats against our records of about 4,000 cats. It works well for us so far: about 64% top-1 on a hard test set of real-world photos (29 of 45 cats), and higher when the match score is high.
We’d like to fine-tune MeowID on our own labeled photos and share the results back with you. We hit one blocker: tools/train.py, tools/validate_training_setup.py and tests/test_training_pipeline.py import cat_recognition.data (build_paired_face_hint_split_dataset), but that module isn’t in the repo or its history. The .gitignore has a data/ line, so it may have been skipped by accident when you published.
Could you add src/cat_recognition/data? If you’d rather not, a description of what it returns would also help. In return we’re glad to share our evaluation results and any fixes we make.
Ryan Tarpey, Flatbush Cats
Thanks for catching this! The data/ rule in .gitignore accidentally excluded the source modules. I’ve restored them and fixed the rule in PR #2.