Skip to content

A neural network coded completely from scratch using only numpy

License

Notifications You must be signed in to change notification settings

t1mtw0/nn_scratch

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Neural Network from scratch

Completely from scratch implementation of a neural network imcluding gradient descent, activation functions, and backpropagation

Usage

python main.py

Implementation

The following are implemented:

  • RELU and Sigmoid activation functions
  • Forward propagation
  • Backpropagation
  • Cross Entropy Loss
  • Derivatives of all the calculated functions

All of the code is batched, and will run with that in mind where the batchs are in the 0th dimension

About

A neural network coded completely from scratch using only numpy

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published