tw.edu.ntu.csie

liblinear

package liblinear

Visibility
  1. Public
  2. All

Type Members

  1. class DataPoint extends Serializable

    DataPoint represents a sparse data point with label.

  2. class Model extends Serializable

    A linear model stores weights and other information.

  3. class Parameter extends Serializable

    Parameter stores the type of solver and user-specified parameters.

  4. class Problem extends Serializable

    A problem stores data points and other necessary information.

  5. class Tron extends AnyRef

    Tron is used to solve an optimization problem by a trust region Newton method.

  6. abstract class TronFunction extends AnyRef

    TronFunction defines necessary methods used for different optimization problems in TRON.

  7. class TronL2SVM extends TronFunction

    TronL2SVM implements TronFunction for L2-regularized L2-loss SVM.

  8. class TronLR extends TronFunction

    TronLR implements TronFunction for L2-regularized Logistic Regression.

Value Members

  1. object Model extends Serializable

  2. object SolverType extends Enumeration

    SolverType defines the names of different solvers.

  3. object SparkLiblinear

    The interface for training liblinear on Spark.

  4. object Utils

    Some useful methods

  5. package rdd

Ungrouped