org.tensorflow.op.core

Class QuantizedConv2DPerChannel<V>

java.lang.Object

org.tensorflow.op.PrimitiveOp

org.tensorflow.op.core.QuantizedConv2DPerChannel<V>

Type Parameters:

V - data type for output() output

All Implemented Interfaces:

Op

public final class QuantizedConv2DPerChannel<V>
extends PrimitiveOp

Computes QuantizedConv2D per channel.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	QuantizedConv2DPerChannel.Options Optional attributes for QuantizedConv2DPerChannel

Field Summary

Fields inherited from class org.tensorflow.op.PrimitiveOp

operation

Method Summary

Modifier and Type	Method and Description
static <V,T,U> QuantizedConv2DPerChannel<V>	create(Scope scope, Operand<T> input, Operand<U> filter, Operand<Float> minInput, Operand<Float> maxInput, Operand<Float> minFilter, Operand<Float> maxFilter, Class<V> outType, List<Long> strides, String padding, QuantizedConv2DPerChannel.Options... options) Factory method to create a class wrapping a new QuantizedConv2DPerChannel operation.
static QuantizedConv2DPerChannel.Options	dilations(List<Long> dilations)
Output<Float>	maxOutput() The maximum value of the final output tensor.
Output<Float>	minOutput() The minimum value of the final output tensor.
Output<V>	output() The output tensor.

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> QuantizedConv2DPerChannel<V> create(Scope scope,
                                                          Operand<T> input,
                                                          Operand<U> filter,
                                                          Operand<Float> minInput,
                                                          Operand<Float> maxInput,
                                                          Operand<Float> minFilter,
                                                          Operand<Float> maxFilter,
                                                          Class<V> outType,
                                                          List<Long> strides,
                                                          String padding,
                                                          QuantizedConv2DPerChannel.Options... options)

Factory method to create a class wrapping a new QuantizedConv2DPerChannel operation.

Parameters:

scope - current scope

input - The original input tensor.

filter - The original filter tensor.

minInput - The minimum value of the input tensor

maxInput - The maximum value of the input tensor.

minFilter - The minimum value of the filter tensor.

maxFilter - The maximum value of the filter tensor.

outType - The quantized type of output tensor that needs to be converted.

strides - list of stride values.

padding -

options - carries optional attributes values

Returns:

a new instance of QuantizedConv2DPerChannel

dilations

public static QuantizedConv2DPerChannel.Options dilations(List<Long> dilations)

Parameters:

dilations - list of dilation values.

output

public Output<V> output()

The output tensor.

minOutput

public Output<Float> minOutput()

The minimum value of the final output tensor.

maxOutput

public Output<Float> maxOutput()

The maximum value of the final output tensor.