SANN - Results - Graphs Tab

Neural Networks

Select the Graphs tab of the SANN - Results dialog box to access the options described here. For information on the options that are common to all tabs, see SANN - Results. Note that for all graph types, you can include cases in the Train, Test, and/or Validation subsets by selecting the appropriate check boxes in the Sample group box. For example, to view a histogram of the outputs for the Validation subset, Select Output in the X-axis box, select the Validation check box in the Sample group, and click the Histograms of X button. Only the predictions for cases in the validation sample will be plotted.

Target variable. Use the Target variable drop-down list to select the target variable (or dependent) to use in plotting graphs. Note that if only one target was specified for the analysis (as is always the case for classification analyses), this option will be disabled.

Case names (2-D and 3-D). Select this check box to include the case names as point labels on 2D and 3D graphs. When case names are not used in the data set, case numbers will be displayed.

X-axis/Y-axis/Z-axis. Use the X-axis, Y-axis, and Z-axis list boxes to select a quantity to plot on the respective axis. Available graph types are dependent on the number of selections you make with these list boxes. For example, if you want to generate a 3D surface plot, you must select a value in each list box. By default, values are selected in the X-axis and Y-axis list box enabling you to create histograms (for the X-axis variables) or 2D scatterplots (for the X-axis and Y-axis variables). Note that if more than one target variable was used in the analysis, you can specify which one to use in the Target variable list box. The available quantities in each list box are given below.

Target. Select Target to plot the target that has been selected in the Target variable list on the selected axis.

Output. Select Output to plot the output (or predicted value) of the target that has been selected in the Dependent variable list on the selected axis.

Residual. Select Residual to plot the residual value (for the selected target variable) on the axis. This is only available for regression type analyses.

Std. residual. Select Std. Residual to plot the standardized residual value (for the selected dependent variable) on the axis. This is only available for regression type analyses.

Abs. residual. Select Abs. Residual to plot the absolute value of the residual (for the selected dependent variable) on the axis. This is only available for regression type analyses.

Sqd. residual. Select Sqd. Residual to plot the squared residual value (for the selected dependent variable) on the axis. This is only available for regression type analyses.

Input variables. Each input variable is listed by name and is also available for selection.

Histograms of X. Click the Histograms of X button to generate a histogram of the quantity selected in the X-axis list box. When there is more than one active network, individual histograms will be generated for each network, when applicable. For example, if you select Residual in the X-axis box, then click Histograms of X, a histogram will be generated for each of the networks in the Active neural networks grid.

X and Y. Click the X and Y button to generate a 2D scatterplot of the variables selected in the X-axis and Y-axis list boxes. When there is more than one active network, a multiple scatterplot will be generated that plots the selected values for all networks, where applicable. For example, if you select Target in the X-axis box and Output in the Y-axis box, then click the X and Y button, only one scatterplot will be generated. It will contain a Target by Output plot for each of the active networks.

X, Y and Z. Click the X, Y and Z button generate a 3D surface plot of the variables selected in the X-axis, Y-axis, and Z-axis list boxes. When there is more than one active network, an individual surface plot will be generated for each network, when applicable. For example, if there are three active networks, a surface plot will be generated for each network.

Training error. Click this button to display a training graph for the active networks. Depending on which options were selected in the Real time training graph display group box ( on the SANN - Custom Neural Network or SANN - Subsampling - Real Time Training Graph tab), the training graph will contain the training, test error, or both.

Note about Graphs and the Missing subset. In SANN, a missing value case is defined as one that has one or more target values missing. This means if, for a data case, all the inputs are intact but one or more target values are missing, then it is labeled as missing (as opposed to train, test, or validation). All missing cases in SANN are grouped together to form a sample called the missing sample. It is used to train the hidden units (radial basis) of RBF networks and to make predictions for MLP or RBF networks. When a data case falls into the missing sample category, only the network outputs are available. In other words, SANN cannot calculate residuals since one or more target is not available.

When creating graphs for target variables or quantities derived from the target variables (e.g. residuals), cases in the missing sample are omitted even if the Missing check box (in the SANN - Results dialog in the Sample group box ) is selected. However, if you create graphs of the inputs, the missing sample will be included (provided the Missing check box is selected).