Package io.github.kirstenali.deepj.tensor

Class CpuNdReduction

java.lang.Object

io.github.kirstenali.deepj.tensor.CpuNdReduction

public final class CpuNdReduction
extends Object

Axis reductions accumulate in FLOAT32 and round compact output only once per result value.

Method Summary

Modifier and Type	Method	Description
static Tensor	mean(Tensor input, int[] axes, boolean keep)
static Tensor	sliceGradient(Tensor gradient, int[] dimensions, int axis, int start, int step)
static Tensor	sum(Tensor input, int[] axes, boolean keep)
static float	total(Tensor input, boolean absolute)

Methods inherited from class java.lang.Object

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

Method Details

sum

public static Tensor sum(Tensor input,
 int[] axes,
 boolean keep)

mean

public static Tensor mean(Tensor input,
 int[] axes,
 boolean keep)

total

public static float total(Tensor input,
 boolean absolute)

sliceGradient

public static Tensor sliceGradient(Tensor gradient,
 int[] dimensions,
 int axis,
 int start,
 int step)