Spaces:
Running
Running
| title: my-ml-platform docs | |
| emoji: ๐ | |
| colorFrom: gray | |
| colorTo: blue | |
| sdk: static | |
| pinned: false | |
| # my-ml-platform | |
| Cloud artefacts for the MLOps serving platform. | |
| ## ONNX models (Hugging Face Hub) | |
| **https://huggingface.co/edu01edu01/my-ml-platform-onnx** | |
| Includes ResNet50 fp32/int8, YOLO26n, DeepLabV3, stub. | |
| ## Live API / Gradio | |
| Docker & Gradio Spaces on free `cpu-basic` now require [HF PRO](https://huggingface.co/pro). | |
| This account does not have PRO, so the interactive Space cannot be hosted here yet. | |
| Options: | |
| 1. Upgrade to PRO, then push `deploy/hf_space` (Docker FastAPI on port 7860). | |
| 2. Run locally: `python deploy/gradio_app.py` (creates a temporary `*.gradio.live` share link). | |
| 3. Point any free host (Render/Fly) at the ONNX repo above. | |
| ## Canonical endpoints (when served) | |
| - `GET /health` | |
| - `GET /models` | |
| - `POST /infer/{name}/{version}` | |
| - `GET /metrics` | |
| - `POST /models/activate/{name}/{version}` | |