doctra-document-parser / requirements.txt
DaVinciCode's picture
modifications done
8364708
# Core dependencies
gradio>=4.0.0,<5
pandas>=2.0.0
numpy>=1.21.0
pillow>=9.0.0
opencv-python>=4.5.0
scikit-image>=0.19.0
torch>=1.12.0
torchvision>=0.13.0
# PDF processing
pdf2image>=1.16.0
pypdfium2>=4.0.0
PyMuPDF>=1.23.0
# OCR and layout detection
paddleocr>=2.6.0
paddlepaddle>=2.4.0
paddlepaddle-gpu>=2.4.0
paddlex>=3.0.0
# VLM providers
openai>=1.0.0
anthropic>=0.3.0
google-generativeai>=0.3.0
httpx>=0.24.0
# Note: google-genai removed due to websockets conflict with Gradio
# The app will work with other VLM providers (OpenAI, Anthropic, etc.)
# Doctra library (install from source)
git+https://github.com/AdemBoukhris457/Doctra.git
# Image processing and restoration
scikit-image>=0.19.3
torchvision
# Utilities
pathlib2>=2.3.0
tqdm>=4.64.0
requests>=2.28.0
beautifulsoup4>=4.11.0
lxml>=4.9.0
openpyxl>=3.0.0
# Hugging Face Spaces specific
huggingface-hub>=0.16.0
transformers>=4.21.0
# Additional dependencies for DocRes
accelerate>=0.20.0
safetensors>=0.3.0