Class Blobs

java.lang.Object
ai.beestream.jitnn.data.Blobs

public final class Blobs extends Object
A synthetic classification set: classes Gaussian clusters in dims dimensions.
  • Field Summary

    Fields
    Modifier and Type
    Field
    Description
    final int
    The number of samples.
    final int
    The number of dimensions.
    final float[]
    The samples [count, dims], row-major.
    final int[]
    The cluster of each sample.
  • Constructor Summary

    Constructors
    Constructor
    Description
    Blobs(int count, int dims, int classes, float spread, long seed)
    count samples around classes centres drawn uniformly in [-2, 2)^dims, with Gaussian noise of standard deviation spread, from seed.
  • Method Summary

    Modifier and Type
    Method
    Description
    void
    batch(int b, int size, float[] xb, int[] yb)
    Copies batch b of size samples into xb and yb.

    Methods inherited from class Object

    clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
  • Field Details

    • x

      public final float[] x
      The samples [count, dims], row-major.
    • y

      public final int[] y
      The cluster of each sample.
    • count

      public final int count
      The number of samples.
    • dims

      public final int dims
      The number of dimensions.
  • Constructor Details

    • Blobs

      public Blobs(int count, int dims, int classes, float spread, long seed)
      count samples around classes centres drawn uniformly in [-2, 2)^dims, with Gaussian noise of standard deviation spread, from seed.
  • Method Details

    • batch

      public void batch(int b, int size, float[] xb, int[] yb)
      Copies batch b of size samples into xb and yb.