Ann

From Eigenvector Research Documentation Wiki
Revision as of 13:09, 15 November 2013 by imported>Donal
Jump to navigation Jump to search

Purpose

Predictions based on Artificial Neural Network (ANN) regression models.

Synopsis

[model] = ann(x,y,options);
[ypred] = ann(x,model,options);
[ypred] = ann(x,y,model,options);
[ypred] = ann(x,y, xt,yt, options);

Description

Build an ANN model from input X and Y block data using the specified number of layers and layer nodes. Alternatively, if a model is passed in ANN makes a Y prediction for an input test X block. The ANN model contains quantities (weights etc) calculated from the calibration data. If a model structure is passed in then these do not have to be re-calculated. The ANN is a Feedforward ANN with Resilient Backpropagation training. See [1] for further details. ANN is implemented using the Encog package [2] provided by Heaton Research, Inc, under the Apache 2.0 license.

Options

options = a structure array with the following fields:

  • display : [ 'off' |{'on'}] Governs display
  • plots: [ {'none'} | 'final' ] governs plotting of results, and
  • waitbar : [ 'off' |{'auto'}| 'on' ] governs use of waitbar during analysis. 'auto' shows waitbar if delay will likely be longer than a reasonable waiting period.
  • anntype : [{1} | 2] Number of hidden layers.
  • nhid1 : [{2}] Number of nodes in first hidden layer.
  • nhid2 : [{2}] Number of nodes in second hidden layer.
  • maxseconds : [{30}] Maximum duration of ANN training.
  • terminalrmse : [0.005] Termination RMSE value for ANN iterations.
  • terminalrmserate : [1.e-6] Termination rate of change of RMSE per 100 iterations.
  • preprocessing: {[] []} preprocessing structures for x and y blocks (see PREPROCESS).
  • compression: [{'none'}| 'pca' | 'pls' ] type of data compression to perform on the x-block prior to calculaing or applying the SVM model. 'pca' uses a simple PCA model to compress the information. 'pls' uses either a pls or plsda model (depending on the svmtype). Compression can make the SVM more stable and less prone to overfitting.
  • compressncomp: [1] Number of latent variables (or principal components to include in the compression model.
  • blockdetails: [ {'standard'} | 'all' ], extent of predictions and residuals included in model, 'standard' = only y-block, 'all' x- and y-blocks.

See Also

modelselector