public static class MaxPoolGradWithArgmax.Inputs<T extends TNumber> extends RawOpInputs<MaxPoolGradWithArgmax<T>>
Modifier and Type | Field and Description |
---|---|
Operand<? extends TNumber> |
argmax
The indices of the maximum values chosen for each output of
max_pool . |
Operand<T> |
grad
4-D with shape
[batch, height, width, channels] . |
boolean |
includeBatchInIndex
Whether to include batch dimension in flattened index of `argmax`.
|
Operand<T> |
input
The original input.
|
long[] |
ksize
The size of the window for each dimension of the input tensor.
|
String |
padding
The type of padding algorithm to use.
|
long[] |
strides
The stride of the sliding window for each dimension of the
input tensor.
|
DataType |
T
The T attribute
|
DataType |
Targmax
The Targmax attribute
|
Constructor and Description |
---|
Inputs(GraphOperation op) |
attributeMetadata, attributeNames, attributes, attributeValue, attributeValues, equals, getOutputs, hashCode, toString
public final Operand<T extends TNumber> grad
[batch, height, width, channels]
. Gradients w.r.t. the
output of max_pool
.public final Operand<? extends TNumber> argmax
max_pool
.public final long[] ksize
public final long[] strides
public final String padding
public final boolean includeBatchInIndex
public final DataType Targmax
public final DataType T
public Inputs(GraphOperation op)
Copyright © 2015–2022. All rights reserved.