InputType

enum InputType : Enum<InputType>

Different Neural Networks were trained on ImageNet dataset with different image preprocessing. The main types of preprocessing widely used in keras.applications are presented in this enumeration.

Entries

TORCH

common

TORCH()

This preprocessing will scale pixels between 0 and 1, then will normalize each channel with respect to the ImageNet dataset.

common

TF()

This preprocessing will scale pixels between -1 and 1, sample-wise.

CAFFE

common

CAFFE()

This preprocessing will convert the images from RGB to BGR, then will zero-center each color channel with respect to the ImageNet dataset, without scaling.

Functions

preprocessing

common

abstract fun preprocessing(channelsLast: Boolean = true): <ERROR CLASS><<ERROR CLASS><FloatArray, <ERROR CLASS>>, <ERROR CLASS><FloatArray, <ERROR CLASS>>>

Returns preprocessing Operation corresponding to this preprocessing type.

Properties

name

common

val name: String

ordinal

common

val ordinal: Int