Apply Kohonen Network

Top  Previous  Next

_bm6Apply Kohonen Network

Use this module to process a data file through a trained neural network to produce the network's classifications or predictions for each pattern in the file.  A file of outputs (the .OUT file) is produced.


Set Output Neuron Values

You can either set the winning neuron to 1, and all others to 0, or show actual neuron values in outputs.  Click on the appropriate check box.  You may want to show actual neuron values to examine which neurons are "runner ups" and how close the runner ups were to the winner.


If you have chosen Vanilla Kohonen (Euclidean distance metric), then the winning neurons have the smallest values.  If you have chosen Normalized, then winning neurons have the largest values.


Use the Run Menu to start processing the data file through the network. Also use this menu to interrupt processing.


Use the File Menu to select an alternate pattern file, view the pattern file, or view the output file.


A prediction will not be made in a row if your data file includes an * in a cell beneath a column labeled I (Input).  (The column label was specified in the Define Inputs/Outputs module.)


File Note: This module defaults to processing the .PAT file, but you can apply the network to any file that is in the NeuroShell 2 file format (the same as  Lotus 1-2-3 .WK1 or Excel 4 .XLS file format) simply by using the File Menu to select a file.  The inputs must be in the same columns in the same order as the .PAT file with which the network was trained.  This module places the network's classifications or predictions into an .OUT file.