Configuring Error Analysis Options

The following choices are available:

• No Error Analysis. No error analysis is performed on the current approximation.
• Separate Data File. This method compares exact and approximate output values for each data point.
  1. Enter the name and path of the data file in the corresponding text box, or click Browse and navigate to the file that contains the approximation data (i.e., the file contains parameter names on the first line and data points on each subsequent line).
  2. Determine how Isight handles the file using one of the following options:

     Static file	Prevents changes to the data in the file. Isight reads the data once and saves it in memory for future use. However, if the contents of the file change, you can instruct Isight to reread the file by clicking Re-read File.
     Dynamic file	Allows changes to the data in the file. Isight reads the data before every initialization. This option is available only in Isight standalone. It is not supported in a SIMULIA Execution Engine environment.

• Cross-Validation. This method selects a subset of points from the main data set, removes each point one at a time, recalculates coefficients, and compares exact and approximate output values at each removed point.
  1. In the text box, type the number of points from the total number of sampling points that you want to use for cross-validation error analysis.
  2. Click Use a fixed random seed for selecting points and specify a seed value to use for the random generator when determining the set of sample points to be executed. This option allows you to reproduce the approximation with the same set of points later.