Package com.jml.neural_network.layers.initilizers

Class GlorotNormal

java.lang.Object
com.jml.neural_network.layers.initilizers.GlorotNormal
All Implemented Interfaces:
Initializer
public class GlorotNormal extends Object implements Initializer
{layer parameter initializer to produce random values from a truncated normal distribution with standard deviation std = sqrt(2 / (fanIn + fanOut)) and mean of 0 where fanIn is the input dimension for the layer and fanOut is the output dimension for the layer.

  • Constructor Summary

    Constructors
    Constructor
    Description
    GlorotNormal()
    Creates a GlorotNormal initializer with specified seed.
    GlorotNormal​(long seed)
    Creates a GlorotNormal initializer with specified seed.

  • Method Summary

    Modifier and Type
    Method
    Description
    linalg.Matrix
    init​(int m, int n)
    Initializes values of a weight/bias matrix as specified by this initializer.

    Methods inherited from class java.lang.Object

    equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

  • Constructor Details

    • GlorotNormal

      public GlorotNormal()
      Creates a GlorotNormal initializer with specified seed.

    • GlorotNormal

      public GlorotNormal​(long seed)
      Creates a GlorotNormal initializer with specified seed.
      Parameters:
      seed - Seed for the random number generator.

  • Method Details

    • init

      public linalg.Matrix init​(int m, int n)
      Initializes values of a weight/bias matrix as specified by this initializer.
      Specified by:
      init in interface Initializer
      Parameters:
      m - Number of rows in the matrix.
      n - Number of columns in the matrix.
      Returns:
      A Matrix initialized with values from the distribution specified by this initializer.