https://github.com/gsteixeira/neuralnetwork-go

A simple feed forward neural network in Go
https://github.com/gsteixeira/neuralnetwork-go

Last synced: about 5 hours ago
JSON representation

A simple feed forward neural network in Go

• Host: GitHub
• URL: https://github.com/gsteixeira/neuralnetwork-go
• Owner: gsteixeira
• Created: 2021-09-30T03:03:51.000Z (about 3 years ago)
• Default Branch: main
• Last Pushed: 2021-10-11T18:30:02.000Z (about 3 years ago)
• Last Synced: 2024-06-21T02:05:15.281Z (5 months ago)
• Language: Go
• Size: 11.7 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# A Neural Network in Go

A feed forward neural network in Go

## usage:

If you just wanna try it:

```shell

    make run

```

## Create a neural network:

Create a network telling the size (nodes) of earch layer.

```go

    nn := NewNeuralNetwork(size_input,

                           size_output,

                           []int {hidden_sizes,})

    // train the network

    nn.Train(inputs, outputs, iteractions)

    // Make predictions

    predicted := nn.Predict(inputs[i])

```

It is possible to use multiple hidden layers. Just pass the size of each one of them in the list:

```go

    // this will create a hidden layer of 3 layers, with sizes 4, 6 and 3.

    nn := NewNeuralNetwork(size_input, size_output,

                           []int {4, 6, 3})

```

## Two implementations:

There are two implementations on this repository:

- neural.go - Object oriented, more flexible and has more features.

- neural_procedural.go - Simple, procedural implentation that uses arrays.