Skip to content

Latest commit

 

History

History
55 lines (41 loc) · 1.08 KB

File metadata and controls

55 lines (41 loc) · 1.08 KB

ONNX Model Export Guide

PyRagix.Net requires ONNX models for embeddings and reranking. Export once from Python.

Prerequisites

pip install optimum-onnx
pip install onnxruntime  # or onnxruntime-gpu for CUDA

Export Models

# Embedding model (sentence-transformers)
optimum-cli export onnx \
  --model sentence-transformers/all-MiniLM-L6-v2 \
  --task feature-extraction \
  pyragix-net-console/Models/embeddings

# Reranker model (cross-encoder)
optimum-cli export onnx \
  --model cross-encoder/ms-marco-MiniLM-L-6-v2 \
  --task text-classification \
  pyragix-net-console/Models/reranker

Verify

Check for model.onnx in each folder:

  • pyragix-net-console/Models/embeddings/model.onnx
  • pyragix-net-console/Models/reranker/model.onnx

Tesseract OCR (Optional)

For image/PDF OCR:

Windows:

# Install from: https://github.com/UB-Mannheim/tesseract/wiki
# Then verify:
tesseract --version

Linux:

sudo apt install tesseract-ocr tesseract-ocr-eng

macOS:

brew install tesseract

Models are gitignored (large files).