Multilayer Perceptron with Deployment (Classification)

Applies Multilayer Perceptron (MLP) neural network architectures to classification problems; the final solution is automatically stored for deployment. MLP neural network architectures with 1, 2, or 3 layers can be specified, for problems with continuous and/or categorical predictors.

Please note: Multilayer perceptrons with many variables and 2 or 3 layers can require a substantial amount of training time, even on fast computers or networks. Always start with simple architectures with few variables and 1 or 2 layers, before applying more complex models.

General

Detail of computed results reported. Detail of computed results; if Minimal detail is requested, summary statistics for the trained network and a graph of the network architecture will be displayed; at the Comprehensive level of detail, sensitivity analysis and descriptive statistics spreadsheets will be displayed; the All results level will display the predictions spreadsheets.

Missing data. Specifies the substitution method for missing data.

Apply memory limit. Use this option to limit the maximum data size that can be processed; note that very large data problems may require significant memory and processing resources; modify the defaults only as needed.

Memory limit. Use this option to set the maximum data size that can be processed.

Save/run network file. By default (Don't save trained networks), the program will simply train the network, report the results, and then discard the trained network. Use the Save network file option to save the trained network in a specific file for future application to other data; use the Run network file option to apply a previously saved network to new data.

Network file name. Specifies the name of the network file to save or run; this option is not applicable if the Save/run network file option was set to Don't save trained networks.

Generate datasource, if N for input less than. Generate a datasource for further analyses with other Data Miner nodes if the input datasource has fewer than k observations, as specified in this edit field; note that parameter k (number of observations) will be evaluated against the number of observations in the input data source, not the number of valid or selected observations.

Units

Number of hidden layers. Specifies the number of hidden layers for the multilayer perceptron, and the number of units in each layer.

Please note: Multilayer perceptrons with many variables and 2 or 3 layers can require a substantial amount of training time, even on fast computers or networks. Always start with simple architectures with few variables and 1 or 2 layers, before applying more complex models.

Number of units layer 1. Specifies the number of units for hidden layer 1.

Number of units layer 2. Specifies the number of units for hidden layer 2; not applicable if the Number of hidden layers was selected to be 1.

Number of units layer 3. Specifies the number of units for hidden layer 3; not applicable if the Number of hidden layers was selected to be 2 or less.

Classification error

Classification error function. Specifies classification error function for output interpretation; applicable to classification-type problems with categorical output variables.

Training

Training phase one. Specifies whether apply phase one of the two-phase training (estimation) procedure; use the Phase one algorithm option to select a specific algorithm for this phase.

Phase one algorithm. Specifies the training phase one algorithm; only applicable if the Training phase one parameter is True.

Phase one epochs. Specifies the epochs (number of iterations) for training phase one; only applicable if the Training phase one parameter is True.

Phase one learning rate. Specifies the learning rate for training phase one; only applicable if the Training phase one parameter is True.

Training phase two. Specifies whether apply phase two of the two-phase training (estimation) procedure; use the Phase two algorithm option to select a specific algorithm for this phase.

Phase two algorithm. Specifies the training phase two algorithm; only applicable if the Training phase two parameter is True.

Phase two epochs. Specifies the epochs (number of iterations) for training phase two; only applicable if the Training phase two parameter is True.

Phase two learning rate. Specifies the learning rate for training phase two; only applicable if the Training phase two parameter is True.

Predicted values

Subset to generate results. Specifies the subset of observations to be used to compute predicted and residual values; only applicable if Comprehensive output or All results are selected as the Detail of computed results reported.

Deployment. Deployment is available if the Statistica installation is licensed for this feature.

Generates C/C++ code. Generates C/C++ code for deployment of predictive model.

Generates SVB code. Generates Statistica Visual Basic code for deployment of predictive model.

Generates PMML code. Generates PMML (Predictive Models Markup Language) code for deployment of predictive model. This code can be used via the Rapid Deployment options to efficiently compute predictions for (score) large data sets.