Predict Real-Time
Predict real-time output value by changing the input value.
Predict real-time output values by changing the input values using the What-If analysis method.
-
Click Next to view the Model Quality values.
The Model Quality and What-If values are displayed.In model quality, a measure of ML model, R2, is displayed for train, cross-validation and test data. The closer it is to 1.0; the better the model is in predicting the known data points. R-square values in the Test column are better indications of the ML model quality but in the absence of a separate test data; R-Square values in the cross-validation column can be used as cross-validation provides a form of testing in the absence of a separate test data. Before you proceed with what-if studies or optimization; you should make sure that the R-square values in test or cross-validation are within acceptable ranges (i.e., greater than 0.7).
-
Change the values of the input values in the What-If
panel to predict real-time output values.
The values of the output are updated by the ML model once the input values are changed.
-
Click in the output to display the
curve.
The curve for the output is displayed.
-
Click to view the simulation model.
Note that this is available for displacements.The simulation model information is displayed.
-
Click to view the 3D Model.
-
Click to view the result contour plot.