From Data to Neural Network to Results.


Creating and training a neural network from your data is a simple five step process. 

Using the neural network to produce results is just as easy.



  Five steps to create and train a neural network from a data file.

  1.   File > New to create a blank grid.  The grid is initialized.

  2.   File > Import to import the data into the grid.  The import dialogs are already partially filled in.  Just a few clicks and the data appears in the grid.

  3.   Action > New Network to create a neural network from the grid.  The neural network dialog is already set to produce the input layer, one hidden layer and the output layer.  One click and a fully connected neural network is created.

  4.   Action > Change Controls to set up the training and validating controls.  The controls dialog is initialized with the default settings.  Any training and validating controls can be adjusted or just press enter to use the defaults.

  5.   Action > Start Learning to train and validate the neural network.  Learning will run until a stop is reached.

  When learning stops the neural network can be used. 

  •   Query > Query to interrogate the neural network interactively.

  •   Query > File Query to interrogate the neural network and create a results file.

Copyright 2002-2008 Neural Planner Software Ltd
Last Updated: 2nd
June 2008