org.tensorflow.op.linalg

Class QuantizedMatMul<V>

java.lang.Object

org.tensorflow.op.PrimitiveOp

org.tensorflow.op.linalg.QuantizedMatMul<V>

Type Parameters:

V - data type for out() output

All Implemented Interfaces:

Op

@Operator(group="linalg")
public final class QuantizedMatMul<V>
extends PrimitiveOp

Perform a quantized matrix multiplication of `a` by the matrix `b`.

The inputs must be two-dimensional matrices and the inner dimension of `a` (after being transposed if `transpose_a` is non-zero) must match the outer dimension of `b` (after being transposed if `transposed_b` is non-zero).

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	QuantizedMatMul.Options Optional attributes for QuantizedMatMul

Field Summary

Fields inherited from class org.tensorflow.op.PrimitiveOp

operation

Method Summary

Modifier and Type	Method and Description
static <V,T,U,W> QuantizedMatMul<V>	create(Scope scope, Operand<T> a, Operand<U> b, Operand<Float> minA, Operand<Float> maxA, Operand<Float> minB, Operand<Float> maxB, Class<V> Toutput, Class<W> Tactivation, QuantizedMatMul.Options... options) Factory method to create a class wrapping a new QuantizedMatMul operation.
Output<Float>	maxOut() The float value that the highest quantized output value represents.
Output<Float>	minOut() The float value that the lowest quantized output value represents.
Output<V>	out()
static QuantizedMatMul.Options	transposeA(Boolean transposeA)
static QuantizedMatMul.Options	transposeB(Boolean transposeB)

Methods inherited from class org.tensorflow.op.PrimitiveOp

equals, hashCode, op, toString

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Method Detail

create

public static <V,T,U,W> QuantizedMatMul<V> create(Scope scope,
                                                  Operand<T> a,
                                                  Operand<U> b,
                                                  Operand<Float> minA,
                                                  Operand<Float> maxA,
                                                  Operand<Float> minB,
                                                  Operand<Float> maxB,
                                                  Class<V> Toutput,
                                                  Class<W> Tactivation,
                                                  QuantizedMatMul.Options... options)

Factory method to create a class wrapping a new QuantizedMatMul operation.

Parameters:

scope - current scope

a - Must be a two-dimensional tensor.

b - Must be a two-dimensional tensor.

minA - The float value that the lowest quantized `a` value represents.

maxA - The float value that the highest quantized `a` value represents.

minB - The float value that the lowest quantized `b` value represents.

maxB - The float value that the highest quantized `b` value represents.

Toutput -

Tactivation - The type of output produced by activation function following this operation.

options - carries optional attributes values

Returns:

a new instance of QuantizedMatMul

transposeA

public static QuantizedMatMul.Options transposeA(Boolean transposeA)

Parameters:

transposeA - If true, `a` is transposed before multiplication.

transposeB

public static QuantizedMatMul.Options transposeB(Boolean transposeB)

Parameters:

transposeB - If true, `b` is transposed before multiplication.

out

public Output<V> out()

minOut

public Output<Float> minOut()

The float value that the lowest quantized output value represents.

maxOut

public Output<Float> maxOut()

The float value that the highest quantized output value represents.