wolffd@0: wolffd@0: wolffd@0: wolffd@0: Netlab Reference Manual demtrain wolffd@0: wolffd@0: wolffd@0: wolffd@0:

demtrain wolffd@0:

wolffd@0:

wolffd@0: Purpose wolffd@0:

wolffd@0: Demonstrate training of MLP network. wolffd@0: wolffd@0:

wolffd@0: Synopsis wolffd@0:

wolffd@0: 
wolffd@0: demtrain
wolffd@0: wolffd@0: wolffd@0:

wolffd@0: Description wolffd@0:

wolffd@0: demtrain brings up a simple GUI to show the training of wolffd@0: an MLP network on classification and regression problems. The user wolffd@0: should load in a dataset (which should be in Netlab format: see wolffd@0: datread), select the output activation function, the wolffd@0: number of cycles and hidden units and then wolffd@0: train the network. The scaled conjugate gradient algorithm is used. wolffd@0: A graph shows the evolution of the error: the value is shown wolffd@0: max(ceil(iterations / 50), 5) cycles. wolffd@0: wolffd@0:

Once the network is trained, it is saved to the file mlptrain.net. wolffd@0: The results can then be viewed as a confusion matrix (for classification wolffd@0: problems) or a plot of output versus target (for regression problems). wolffd@0: wolffd@0:

wolffd@0: See Also wolffd@0:

wolffd@0: confmat, datread, mlp, netopt, scg
wolffd@0: Pages: wolffd@0: Index wolffd@0:
wolffd@0:

Copyright (c) Ian T Nabney (1996-9) wolffd@0: wolffd@0: wolffd@0: wolffd@0: