| --- |
| title: HyperView-Jaguar-ReID |
| emoji: 🐆 |
| colorFrom: yellow |
| colorTo: green |
| sdk: docker |
| app_port: 7860 |
| pinned: false |
| --- |
| |
| # HyperView - Jaguar Re-ID (MegaDescriptor + Sphere) |
|
|
| This Space runs the Jaguar Re-ID dataset through the MegaDescriptor timm |
| backbone and renders the result with HyperView's spherical 3D layout. |
|
|
| This example now installs the released `hyperview[ml]==0.3.1` package from |
| PyPI, which includes the `timm-image` provider and spherical 3D layout support |
| required by this demo. |
|
|
| This demo uses: |
|
|
| - Hugging Face dataset `hyper3labs/jaguar-re-id` |
| - Config `default` |
| - Split `train` |
| - Image field `image` |
| - Label field `label` |
| - Sample count `200` |
| - Embedding model `hf-hub:BVRA/MegaDescriptor-L-384` |
| - Layout `spherical` (3D) |
|
|
| ## Build model |
|
|
| The Dockerfile precomputes the dataset, embeddings, and layout during image |
| build so the runtime container only needs to launch HyperView. |
|
|
| Because MegaDescriptor inference runs during Docker build on CPU, this Space |
| keeps the sample count modest and uses a smaller batch size than the local demo |
| script to stay within typical Hugging Face build limits. |
|
|
| ## Reuse this example |
|
|
| If you need a simple starter, copy `spaces/imagenette-clip-hycoclip` first. |
| If your own Space needs a timm backbone or a spherical 3D layout, copy this |
| folder instead and change the constants block at the top of [demo.py](demo.py). |
|
|
| When contributing your own Space back to this repository, add a row to the |
| community table in the root `README.md` and include your Hugging Face Space ID |
| in the pull request description. |
|
|
| ## Deploy source |
|
|
| This folder is synchronized to Hugging Face Spaces by GitHub Actions from the |
| `hyperview-spaces` deployment repository. |
