Skip to content

Create a simple neural network and testing with a simple example

Notifications You must be signed in to change notification settings

FzFStormZ/NeuralNetwork

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 

Repository files navigation

NeuralNetwork

Create a simple neural network and testing with a simple example

For what

There is the minimum code to begin with easy examples which use simple data such as math operations (XOR, NAND, ...)

How to use

  • There is inputs and outputs in main class which are data to your examples
  • After training with these data, there is a testing phase in order to test with new data to know if your neural network works correctly

How to work Neural Network

Before training

  • We have some inputs and the results of these inputs such as :

          |Inputs |Outputs |
          |-------|--------|
          |0, 0, 1| 0      |
          |1, 1, 1| 1      |
          |1, 0, 1| 1      |
          |0, 1, 1| 0      |
    
          > A row represent 3 inputs (3 neurons in input layer) and 1 output (1 neuron in output layer)
    
  • Moreover, we have some weights for every synapse choosed randomly

During training

  • We multiply the matrix of inputs values and the matrix of their weights

  • A bias is added during this sum : ∑ (values + weights) + bias (from 1 to n neurons) : Sum things

  • After that, we send this matrix of outputs results in sigmoid function : Activation Function

    • Sigmoid Function : SigmoidFunction
      • x is the result.
  • Backpropagation :

    • We compute the error

!! NOT FINISHED YET !!

About

Create a simple neural network and testing with a simple example

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages