Spaces:
Running on Zero
Running on Zero
|
Download README.md from RISHIVEL/RAP_DocLayout_DetectionD: direct link, hf CLI and curl.
- Browser
- Download file 716 Bytes
-
https://huggingface.co/spaces/RISHIVEL/RAP_DocLayout_DetectionD/resolve/main/README.md
- Command line
-
hf download hf://spaces/RISHIVEL/RAP_DocLayout_DetectionD/README.md
-
curl -L -o README.md https://huggingface.co/spaces/RISHIVEL/RAP_DocLayout_DetectionD/resolve/main/README.md
716 Bytes
| title: Document Layout Detection | |
| emoji: ๐ | |
| colorFrom: blue | |
| colorTo: gray | |
| sdk: gradio | |
| app_file: gradio_app.py | |
| pinned: false | |
| # Constrained Document Layout Detection | |
| RT-DETR fine-tuned on DocLayNet, plus a hand-written reasoning layer for | |
| natural-language questions about a page's layout. | |
| Full source, training scripts, and the FastAPI API this demo shares its | |
| code with: see the main repo. | |
| ## Space secrets needed | |
| - `GROQ_API_KEY` - for the Ask tab's reasoning layer | |
| - `MODEL_URL` - direct download link for the trained weights (e.g. a | |
| GitHub Release asset). Downloaded automatically on first run if | |
| `MODEL_PATH` doesn't already exist - no need to push the weights file | |
| into the Space's git repo. | |