Skip to content

avr34/NeuralNet

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

NeuralNet

NeuralNet is a C library for training Feed-Forward Neural Networks. Currently still under development, but it can train to >95% accuracy using the ReLU activation function.

Installation

Clone the repository wherever you like, then call make.

Usage

Initialize a struct for all the training data, and a struct for the network.

int numLayers = 4;
int numNodes[] = {2, 3, 3, 1};
fnnData * trainingData = load_fnnData(numInputs, numOutputs, "filename.csv");
fnnNet * neuralNet = initNet(numLayers, numNodes);

Then train the network using trainNet, and print out the resulting weights.

int numEpochs = 3000;
trainNet(trainingData, neuralNet, numEpochs);
printWeights(neuralNet);

About

Simple Feed-Forward Neural Network library for C/C++.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published