Occular-OCR Wiki examples — release 0.5.1

Extract all files into one folder. Keep inputs beside the notebooks.
From that folder in a fresh Python 3.11/3.12 environment:

python -m pip install "occular-ocr==0.5.1" jupyterlab matplotlib
python -m jupyterlab

Open Quickstart.ipynb first and run every cell in order. First use needs
internet for model downloads. Other notebooks may download optional models.
GPU.ipynb additionally needs the GPU extra and a compatible NVIDIA driver:
python -m pip install "occular-ocr[gpu]==0.5.1"

The visible code uses ordinary Matplotlib and print. Excerpts are selected
explicitly; actual OCR mistakes are retained. Saved notebook output came from
real executions, not corrected transcription. Timings and local model files
will differ on your machine. A package pin does not pin upstream model files.

Synthetic/public sample documents only. Export-results.ipynb writes invoice.json
and table.csv into this folder; later runs can overwrite them.

Page explanations: https://github.com/Bodhi42/Occular-ocr/wiki
