Last updated: Jan 17, 2024
With the Linear Support Vector Machine (LSVM) node, you can classify data into one
of two groups without overfitting. LSVM is linear and works well with wide data sets, such as those
with a very large number of records.
lsvmnode Properties |
Values | Property description |
|---|---|---|
intercept
|
flag | Includes the intercept in the model. Default value is True. |
target_order
|
Ascending
Descending |
Specifies the sorting order for the categorical target. Ignored for continuous targets.
Default is Ascending. |
precision
|
number | Used only if measurement level of target field is Continuous. Specifies the
parameter related to the sensitiveness of the loss for regression. Minimum is 0 and
there is no maximum. Default value is 0.1. |
exclude_missing_values
|
flag | When True, a record is excluded if any single value is missing. The default
value is False. |
penalty_function
|
L1
L2
|
Specifies the type of penalty function used. The default value is
L2. |
lambda |
number | Penalty (regularization) parameter. |
calculate_variable_importance |
flag | For models that produce an appropriate measure of importance, this option displays a chart that indicates the relative importance of each predictor in estimating the model. Note that variable importance may take longer to calculate for some models, particularly when working with large datasets, and is off by default for some models as a result. Variable importance is not available for decision list models. |