LinearOptimizerType enum
Linear optimization types
The whole process of learning linear models confines to just finding coefficients of features of predicting line/hyperplane. There are several algorithms to find the coefficients that minimize a cost function best.
Constructors
- LinearOptimizerType()
-
const
Values
- gradient → const LinearOptimizerType
-
Original gradient descent/ascent optimization, only L2 regularization is applicable while optimizing a function using this method
- coordinate → const LinearOptimizerType
-
Original coordinate descent optimization, only L1 regularization is applicable while optimizing a function using this method
- closedForm → const LinearOptimizerType
-
Analytical solution of the linear regression problem. The solution based on finding the inverse of the feature matrix using LU decomposition
- newton → const LinearOptimizerType
-
The Newton method
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<
LinearOptimizerType> - A constant List of the values in this enum, in order of their declaration.