README.md

metadata

title: OCRapi
emoji: 🐠
colorFrom: gray
colorTo: indigo
sdk: gradio
sdk_version: 5.49.1
app_file: app.py
pinned: false

PDF OCR Space (PaddleOCR + PyMuPDF)

Gradio Space that extracts text from uploaded PDFs. Each page is rendered with PyMuPDF and recognized by PaddleOCR.

Features

• Upload a PDF and extract text
• Adjustable DPI and optional page limit
• Multi-language OCR (via PaddleOCR language packs)
• JSON output including bboxes and confidences per page

App Structure

• app.py: Gradio UI and prediction logic
• requirements.txt: Python dependencies
• runtime.txt: Python runtime for Hugging Face Spaces

Deploy on Hugging Face Spaces

1. Create a new Space (SDK: Gradio, Python).
2. Upload all files in this repo: app.py, requirements.txt, runtime.txt, README.md.
3. First build will download OCR models; subsequent runs are cached.

Run Locally

pip install -r requirements.txt
python app.py

Then open http://127.0.0.1:7860.

API Usage

Using gradio_client in Python:

from gradio_client import Client

client = Client("<your-username>/<your-space-name>")
result = client.predict(
    pdf_file=("file", "./sample.pdf"),
    dpi=170,
    max_pages=None,
    lang="en",
    api_name="predict"
)

text, json_payload = result
print(text)
print(json_payload)

Using Spaces Inference API (curl):

curl -s -X POST \
  -F "data=@sample.pdf" \
  -F "data=170" \
  -F "data=" \
  -F "data=en" \
  https://<your-username>-<your-space-name>.hf.space/run/predict

Notes:

• Increase DPI for small text, but it will be slower.
• Choose the appropriate language model; defaults to en.

License

MIT

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference