https://github.com/rahul-404/handwritten-digit-recognition-mnist

https://github.com/rahul-404/handwritten-digit-recognition-mnist

Last synced: about 2 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/rahul-404/handwritten-digit-recognition-mnist
• Owner: Rahul-404
• License: mit
• Created: 2021-06-10T15:46:57.000Z (almost 4 years ago)
• Default Branch: main
• Last Pushed: 2021-06-14T11:52:35.000Z (almost 4 years ago)
• Last Synced: 2025-02-01T22:13:59.652Z (4 months ago)
• Language: Python
• Size: 24.6 MB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Handwritten-digit-recognition-MNIST

Handwritten Digit Recognition using Convolutional Neural Networks in Python with Keras

## MNIST dataset:

MNIST is a collection of handwritten digits from 0-9. Image of size 28 X 28

![MNIST](mnist-sample.png)

## Description:

This is a 8 layers Sequential Convolutional Neural Network for digits recognition trained on MNIST dataset. I choosed to build it with keras API (Tensorflow backend) which is very intuitive.

It achieved 84.57% of accuracy with this CNN trained on a CPU, which took me about a 5-10minute. If you dont have a GPU powered machine it might take a little longer, you can try reducing the epochs (steps) to reduce computation.

![Python](https://img.shields.io/badge/python-3.x-orange.svg)

![Numpy](https://img.shields.io/badge/Numpy-1.20.3-blue)

![Keras](https://img.shields.io/badge/Keras-2.4.0-red)

![Tensorflow](https://img.shields.io/badge/Tensorflow-2.0-orange)

![Type](https://img.shields.io/badge/Deep-Learning-red.svg) ![Type](https://img.shields.io/badge/Type-Supervised-yellow.svg)

![Status](https://img.shields.io/badge/Status-Completed-yellowgreen.svg)

## Code Requirements:

python 3.x with following modules installed

1. numpy

2. keras

3. opencv2

## Execution:

`pip install -r requirements.txt`

`python model.py`

`python run.py`

![Output](i1.png)

![Output](i2.png)

## Built With

* [Keras](https://keras.io/about/) - Keras is a deep learning API

* [Opencv2](https://opencv.org/) - open-source computer vision library

## Purpose

The purpose of this project was to gain introductory exposure to Deep Learning Classification concepts. The project makes heavy use of Keras(tensorflow in backend) and Opencv2 Libraries.

## Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

Please make sure to update tests as appropriate.

## License

[MIT](https://github.com/Rahul-404/Handwritten-digit-recognition-MNIST/blob/main/LICENSE)