GlobalMaxPool2D

class GlobalMaxPool2D(name: String) : Layer

Global max pooling operation for 2D spatial data (e.g. images).

Downsamples the input by taking the maximum value over spatial dimensions.

Since

0.3

Constructors

GlobalMaxPool2D
Link copied to clipboard
fun GlobalMaxPool2D(name: String = "")

Functions

build
Link copied to clipboard
open override fun build(tf: Ops, input: Operand<Float>, isTraining: Operand<Boolean>, numberOfLosses: Operand<Float>?): Operand<Float>
open fun build(tf: Ops, input: List<Operand<Float>>, isTraining: Operand<Boolean>, numberOfLosses: Operand<Float>?): Operand<Float>

Extend this function to define variables in the layer and compute layer output.

invoke
Link copied to clipboard
operator fun invoke(vararg layers: Layer): Layer

Important part of functional API. It takes layers as input and saves them to the inboundLayers of the given layer.

toString
Link copied to clipboard
open override fun toString(): String

Properties

hasActivation
Link copied to clipboard
open override val hasActivation: Boolean

Returns True, if layer has internal activation function.

inboundLayers
Link copied to clipboard
var inboundLayers: MutableList<Layer>

Returns inbound layers.

name
Link copied to clipboard
var name: String

Layer name. A new name is generated during model compilation when provided name is empty.

outboundLayers
Link copied to clipboard
var outboundLayers: MutableList<Layer>

Returns outbound layers.

outputShape
Link copied to clipboard
lateinit var outputShape: TensorShape

Output data tensor shape.

parentModel
Link copied to clipboard
var parentModel: GraphTrainableModel? = null

Model where this layer is used.