Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

99 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Neural Network

An Neural Network Libary, made completely from scratch using numpy and python, originally based on this article, currently supports:

  • loading, and automatic saving, of trained models
  • Fully Connected Layers
  • Convolutional Layers
  • Softmax
  • Sigmoid Activation
  • Tanh Activation
  • ReLU Activation
  • Custom layer types (inherits Layer)
  • Custom Activations and Loss functions (inherits NNFunction)
  • Mean Squared Error loss
  • Categorical Cross Entropy loss

TODO

  • update pooling layer for multiple color channels
  • convert modules to C for speed?
  • UNets
  • Diffusion Models

About

Machine Learning from scratch

Topics

Resources

Stars

2 stars

Watchers

1 watching

Forks

Contributors

Languages