https://github.com/msikorski93/persian-digits-lenet-5-scikit-learn

Image recognition on Persian handwritten digits with LeNet-5 neural network and scikit-learn models.
https://github.com/msikorski93/persian-digits-lenet-5-scikit-learn

classification gradient-boosting hard-voting-classifier image-recognition k-nearest-neighbors lenet-5 lightgbm persian-digits tensorflow

Last synced: 7 months ago
JSON representation

Image recognition on Persian handwritten digits with LeNet-5 neural network and scikit-learn models.

• Host: GitHub
• URL: https://github.com/msikorski93/persian-digits-lenet-5-scikit-learn
• Owner: msikorski93
• Created: 2023-11-30T13:42:08.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2024-10-22T16:39:27.000Z (12 months ago)
• Last Synced: 2025-02-26T15:17:32.747Z (7 months ago)
• Topics: classification, gradient-boosting, hard-voting-classifier, image-recognition, k-nearest-neighbors, lenet-5, lightgbm, persian-digits, tensorflow
• Language: Jupyter Notebook
• Homepage:
• Size: 432 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Persian-Digits-LeNet-5

![ alt text ](https://img.shields.io/badge/license-MIT-green?style=&logo=)

![ alt text ](https://img.shields.io/badge/-Jupyter-F37626?logo=Jupyter&logoColor=white)

![ alt text ](https://img.shields.io/badge/-Keras-D00000?logo=Keras&logoColor=white)

![ alt text ](https://img.shields.io/badge/-TensorFlow-FF6F00?logo=TensorFlow&logoColor=white)

![ alt text ](https://img.shields.io/badge/-NumPy-013243?logo=Numpy&logoColor=white)

![ alt text ](https://img.shields.io/badge/-scikit--learn-F7931E?logo=scikitlearn&logoColor=white)

New update for October 2024. Persian digits were predicted with `scikit-learn` models and hard voting. The top estimator was the histogram-based gradient boosting algorithm with accuracy 0.9972 ± 0.0003 achieved in cross-validation for test dataset.

Image recognition on handwritten Persian digits performed with LeNet-5 architecture. Accuracy achieved on test dataset: 0.9964.