Package org.jetbrains.kotlinx.dl.api.core.layer.pooling

Types

AvgPool1D

class AvgPool1D(poolSize: IntArray, strides: IntArray, padding: ConvPadding, name: String) : Layer

Average pooling operation for 1D temporal data (e.g. audio, time-series).

AvgPool2D

class AvgPool2D(poolSize: IntArray, strides: IntArray, padding: ConvPadding, name: String) : Layer

Average pooling layer for 2D inputs (e.g. images).

AvgPool3D

class AvgPool3D(poolSize: IntArray, strides: IntArray, padding: ConvPadding, name: String) : Layer

Average pooling operation for 3D data (e.g. video, spatio-temporal).

GlobalAvgPool1D

class GlobalAvgPool1D(name: String) : Layer

Global average pooling operation for temporal data. NOTE: Works with tensors which must have rank 3 (batch, steps, features). Input shape: 3D tensor with shape (batch_size, steps, features). Output shape: 2D tensor with shape (batch_size, features).

GlobalAvgPool2D

class GlobalAvgPool2D(name: String) : Layer

Global average pooling operation for 2D data (images and so on).

GlobalAvgPool3D

class GlobalAvgPool3D(name: String) : Layer

Global Average pooling operation for 3D data.

GlobalMaxPool1D

class GlobalMaxPool1D(name: String) : Layer

Global max pooling operation for 1D temporal data (e.g. audio, timeseries, etc.).

GlobalMaxPool2D

class GlobalMaxPool2D(name: String) : Layer

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

GlobalMaxPool3D

class GlobalMaxPool3D(name: String) : Layer

Global max pooling operation for 3D data (e.g. videos, spatio-temporal).

MaxPool1D

class MaxPool1D(poolSize: IntArray, strides: IntArray, padding: ConvPadding, name: String) : Layer

Max pooling operation for 1D temporal data (e.g. audio, timeseries).

MaxPool2D

class MaxPool2D(poolSize: IntArray, strides: IntArray, padding: ConvPadding, name: String) : Layer

Max pooling layer for 2D inputs (e.g. images).

MaxPool3D

class MaxPool3D(poolSize: IntArray, strides: IntArray, padding: ConvPadding, name: String) : Layer

Max pooling operation for 3D data (spatial or spatio-temporal). NOTE: Works with tensors which must have rank 5 (batch, depth, height, width, channels).