Skip to content

RowlandOti/NeuralNetwork-Framework

Repository files navigation

A Neural network framework

  • An implementation of a simple neural network_
  • Done in Python and Numpy, with no external machine learning framework used_

Neuron Network

The purpose of this project was to build a simple and lighweight neuralNetwork framework. It covers:

Features of a Neural Network Included:
  • Graphs

Topological Sort of Graph

  • Forward and Back Propagation

Back Propagation

  • Linear Transform
  • Sigmoid Activation Function

Sigmoid Activation Function

  • Cost
  • Gradient Descent
  • Stochastic Gradient Descent
  • Learning and Loss

Gradient Descent Convergence

About

⚓ A Simple and Lightweight NeuralNetwork Framework

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published