RegularizationType enum
Regularization types
Machine learning linear models are prone to overfitting, or, in other words, they may lose generalization ability in their learning process: overfitted models show quite high quality on training data, but in the same time they perform very badly on previously unseen data.
The main reason of that is uncontrolled growth of linear model coefficients. To avoid this, it is needed to measure a magnitude of coefficients vector and consider it during the model's learning.
Constructors
- RegularizationType()
-
const
Values
- L1 → const RegularizationType
-
uses Manhattan norm of a vector to calculate magnitude of learned coefficients. Applicable for LinearOptimizerType.coordinate
- L2 → const RegularizationType
-
uses Euclidean norm of a vector to calculate magnitude of learned coefficients. Applicable for LinearOptimizerType.gradient
Properties
Methods
-
noSuchMethod(
Invocation invocation) → dynamic -
Invoked when a nonexistent method or property is accessed.
inherited
-
toString(
) → String -
A string representation of this object.
inherited
Operators
-
operator ==(
Object other) → bool -
The equality operator.
inherited
Constants
-
values
→ const List<
RegularizationType> - A constant List of the values in this enum, in order of their declaration.