cat-handwriting-recognizer 1.2.5

A lightweight handwritten character recognition module using template-based KNN.

Cat Handwriting Recognizer

A simple Python module for recognizing handwritten using machine learning. Designed for easy integration and efficient preprocessing.

Features

• Preprocess handwritten images
• Extract simple features for recognition
• K-Nearest Neighbors (KNN) prediction
• Async support for building templates
• Timeout mechanism for long-running operations
• Save and load models as .npz files

Installation

Install from PyPI:

pip install cat-handwriting-recognizer

Install from GitHub:

pip install git+https://github.com/littlecommandcat/cat_handwriting_recognizer.git

Quick Example

from cat_handwriting_recognizer import HandwritingRecognizer  
import asyncio

recognizer = HandwritingRecognizer(timeout=-1)

# Build templates from data
asyncio.run(recognizer.build_all_templates("directory")) # Your data directory

# Save model
recognizer.save_model("model.npz") # Save model file(.npz)

# Load model
recognizer.load_model("model.npz") # Load model file(.npz)

# Predict a picture
result = asyncio.run(recognizer.predict("file.png")) # Use model to predict png/jpg/jpeg file
print(result)

Requirements

• Python 3.10+
• NumPy
• Pillow

License

MIT License

Author

littlecommandcat