https://github.com/abdulaziz-hassan/mnist_classification

A Deep Neural Network for MNIST Classification
https://github.com/abdulaziz-hassan/mnist_classification

Last synced: 6 days ago
JSON representation

A Deep Neural Network for MNIST Classification

• Host: GitHub
• URL: https://github.com/abdulaziz-hassan/mnist_classification
• Owner: Abdulaziz-Hassan
• Created: 2021-09-26T22:21:09.000Z (over 3 years ago)
• Default Branch: main
• Last Pushed: 2021-09-29T22:41:59.000Z (over 3 years ago)
• Last Synced: 2024-11-19T20:52:30.712Z (2 months ago)
• Language: Jupyter Notebook
• Size: 504 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# MNIST dataset classification with a Deep Neural Network

## About the data

* The MNIST database (Modified National Institute of Standards and Technology database) is a large database of handwritten digits that is commonly used for training various image processing systems. The database is also widely used for training and testing in the field of machine learning.

* The MNIST database contains 60,000 training images and 10,000 testing images.

* Data taken from tensorflow.keras.datasets module.

![MNIST](https://miro.medium.com/max/2000/1*v6Nn_zj8W0qsnk9Cc6Di_g.png)

## Model

A Deep Neural Network with 2 hidden layers of size 100 with 'relu' as their activation functions and a output layer of size 10 with 'softmax' as the activation function.

![NN](https://3.bp.blogspot.com/-mDyzBzA4btg/V4_Z0f2mc7I/AAAAAAAAE3M/dtU8hT661fQWtnRC_JvIH_4qifQomZ4PACLcB/s1600/MNIST_neuralnet_image.png)

## Evaluation Mertrics

* Precision

* Recall

* F1-score

* Confusion Matrix

## Overall Model Accuracy

* 97.14%