# WenmuZhou/PytorchOCR

基于Pytorch的OCR工具库，支持常用的文字检测和识别算法

Repository: https://github.com/WenmuZhou/PytorchOCR
Canonical: https://ross.abutalabs.com/products/pytorchocr
Language: Python
License Family: other
Last push: 2026-01-04T02:40:44+00:00

## Health v2 (maintenance only)
Score: 59/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 60, release rhythm 35, longevity 100
- inputs: {"age_days": 2302, "days_push": 241, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1523, forks 316 (observed 2026-08-28T04:04:58.140811+00:00)

## What it is
A PyTorch-based OCR toolkit that ports PaddleOCR models to PyTorch, supporting common text detection and recognition algorithms like the PP-OCR series. It includes training, evaluation, inference, and model export/prediction workflows with verified numerical alignment against the original PaddleOCR weights.

## Use cases
- run OCR on images with pytorch instead of paddlepaddle
- detect and recognize text in photos
- convert PaddleOCR models to PyTorch
- train a custom text recognition model
- extract text from scanned documents
- deploy PP-OCRv4 models in a PyTorch pipeline

## When to choose
- you need PaddleOCR-quality OCR but your stack is PyTorch
- you want to fine-tune or retrain OCR models in PyTorch
- you need verified model parity with PaddleOCR weights

## When to avoid
- you just want a ready-made OCR CLI with no training (use PaddleOCR or Tesseract)
- you need a license-cleared dependency (this repo has no license)
- you need models not yet in the verified alignment list

## Facets
- artifact type: library
- maturity: active
- function: ocr, machine-learning, deep-learning, image-processing, computer-vision
- domain: computer-vision, image-processing, deep-learning, pdf
- platform: python, cross-platform
- tags: pytorch, text-detection, text-recognition, paddleocr, model-conversion

## Member repositories
- WenmuZhou/PytorchOCR (main) score 59

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:58.140811+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:31:44.095084+00:00, confidence not recorded.
  - readme: https://github.com/WenmuZhou/PytorchOCR (fetched 2026-08-28T04:04:58.140811+00:00, sha c99498973846)
- Data as of 2026-08-30T08:39:29.467469+00:00.