28 lines
634 B
Plaintext
28 lines
634 B
Plaintext
# Core OCR
|
|
paddleocr>=2.7.0
|
|
# PaddlePaddle (choose CPU or GPU build appropriate for your system)
|
|
# For CPU-only install: `pip install paddlepaddle` or follow the official install guide
|
|
#paddlepaddle>=2.5.0
|
|
|
|
# PDF -> image
|
|
pdf2image>=1.16.0
|
|
Pillow>=10.0.0
|
|
|
|
# Database
|
|
pymongo>=4.4.0
|
|
|
|
# Utilities
|
|
python-dotenv>=1.0.0
|
|
rich>=13.0.0
|
|
tqdm>=4.65.0
|
|
|
|
# Testing / linting (optional)
|
|
pytest>=8.0.0
|
|
mypy>=1.5.0
|
|
|
|
# YAML config parsing
|
|
PyYAML>=6.0
|
|
|
|
# Notes:
|
|
# - `pdf2image` requires the `poppler` system package (e.g. `brew install poppler` on macOS).
|
|
# - Select the appropriate `paddlepaddle` wheel for your platform (CPU vs GPU, macOS vs Linux). |