Skip to content

Achieved 96% training data accuracy by implementing a self-developed multi-layer perceptron (MLP) neural network with feed-forward and back-propagation algorithms, without utilizing machine learning libraries like PyTorch or TensorFlow.

Notifications You must be signed in to change notification settings

wallinslax/handwritten-digits-classifier

Repository files navigation

Handwritten Digits Classifier

Sample Digits from MNIST dataset

Description

This project achieved 96% training data accuracy by implementing a self-developed multi-layer perceptron (MLP) neural network with feed-forward and back-propagation algorithms, without utilizing machine learning libraries like PyTorch or TensorFlow.

Getting Started

git clone https://github.com/wallinslax/handwritten-digits-classifier.git
cd data
python mnist_csv3.py
cd ..
python mlpForHandwrritenDigit.py

About

Achieved 96% training data accuracy by implementing a self-developed multi-layer perceptron (MLP) neural network with feed-forward and back-propagation algorithms, without utilizing machine learning libraries like PyTorch or TensorFlow.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages