ocr-man / README.md
Equan's picture
Rename Space to OCR Man
9572e2d
|
Raw
History Blame Contribute Delete
1.14 kB
---
title: OCR Man
emoji: ๐Ÿ†
colorFrom: gray
colorTo: gray
sdk: gradio
sdk_version: 6.19.0
python_version: '3.10'
app_file: app.py
pinned: true
license: mit
hardware: zero-gpu
models:
- baidu/Unlimited-OCR
---
# OCR Man
A Hugging Face Space for [`baidu/Unlimited-OCR`](https://huggingface.co/baidu/Unlimited-OCR), supporting image and PDF document parsing.
## Notes
- This Space is configured for Hugging Face ZeroGPU (`hardware: zero-gpu`).
- The app loads `baidu/Unlimited-OCR` with `trust_remote_code=True`, so deploy it only in an environment where you trust the model repository code.
- The backend installs the model-required `torch`, `torchvision`, and `transformers` versions at runtime to avoid build-time dependency conflicts with newer Gradio/Hugging Face Hub packages.
## Deploy
Create a new Hugging Face Space with the Gradio SDK, then push these files to the Space repository:
```bash
git remote add space https://huggingface.co/spaces/<your-username-or-org>/<space-name>
git push space main
```
If you use private/gated resources later, set the required secrets in the Space settings rather than committing them.