Use this to specify the results to be saved from training a support vector machine in Genstat data structures.
- After selecting the appropriate boxes, type the names for the identifiers of the data structures into the corresponding In: fields.
Save
| Groups | Factor | A factor with the fitted values from the support vector machine |
| Error rates | Scalar, variate or matrix | The error rates for the combinations of the penalty, gamma or gamma specified. If the penalty, gamma and nu each have only a single value a scalar is returned, if one of these has multiple values, a variate is returned, otherwise a matrix is returned |
| Optimum penalty | Scalar | The value of the penalty which gives the smallest error rate when multiple values of the penalty have been specified. |
| Optimum gamma | Scalar | The value of gamma which gives the smallest error rate when multiple values of gamma have been specified |
| Optimum nu | Scalar | The value of nu which gives the smallest error rate when multiple values of nu have been specified |
| Optimum error rates | Scalar | The optimum error rate when multiple values of the penalty, gamma or nu have been specified |
Display in spreadsheet
The saved results will be displayed within a new spreadsheet.
See also
- Train Support Vector Machine menu
- Options for choosing which results to display and options
- Predict from Support Vector Machine for predicting from the fitted support vector machine
- Discriminant analysis menu
- SVMFIT procedure
- SVMPREDICT procedure