Inception family of models
This is the API reference for the Inception family of models supported by Metalhead.jl.
The higher-level model constructors
Metalhead.GoogLeNet
— TypeGoogLeNet(; pretrain::Bool = false, inchannels::Integer = 3, nclasses::Integer = 1000)
Create an Inception-v1 model (commonly referred to as GoogLeNet
) (reference).
Arguments
pretrain
: set totrue
to load the model with pre-trained weights for ImageNetnclasses
: the number of output classesbatchnorm
: set totrue
to use batch normalization after each convolutionbias
: set totrue
to use bias in the convolution layers
GoogLeNet
does not currently support pretrained weights.
See also Metalhead.googlenet
.
Metalhead.Inceptionv3
— TypeInceptionv3(; pretrain::Bool = false, inchannels::Integer = 3, nclasses::Integer = 1000)
Create an Inception-v3 model (reference).
Arguments
pretrain
: set totrue
to load the pre-trained weights for ImageNetinchannels
: number of input channelsnclasses
: the number of output classes
Inceptionv3
does not currently support pretrained weights.
See also Metalhead.inceptionv3
.
Metalhead.Inceptionv4
— TypeInceptionv4(; pretrain::Bool = false, inchannels::Integer = 3,
nclasses::Integer = 1000)
Creates an Inceptionv4 model. (reference)
Arguments
pretrain
: set totrue
to load the pre-trained weights for ImageNetinchannels
: number of input channels.nclasses
: the number of output classes.
Inceptionv4
does not currently support pretrained weights.
See also Metalhead.inceptionv4
.
Metalhead.InceptionResNetv2
— TypeInceptionResNetv2(; pretrain::Bool = false, inchannels::Integer = 3,
nclasses::Integer = 1000)
Creates an InceptionResNetv2 model. (reference)
Arguments
pretrain
: set totrue
to load the pre-trained weights for ImageNetinchannels
: number of input channels.nclasses
: the number of output classes.
InceptionResNetv2
does not currently support pretrained weights.
See also Metalhead.inceptionresnetv2
.
Metalhead.Xception
— TypeXception(; pretrain::Bool = false, inchannels::Integer = 3, nclasses::Integer = 1000)
Creates an Xception model. (reference)
Arguments
pretrain
: set totrue
to load the pre-trained weights for ImageNet.inchannels
: number of input channels.nclasses
: the number of output classes.
Xception
does not currently support pretrained weights.
See also Metalhead.xception
.
The mid-level functions
Metalhead.googlenet
— Functiongooglenet(; dropout_prob = 0.4, inchannels::Integer = 3, nclasses::Integer = 1000)
Create an Inception-v1 model (commonly referred to as GoogLeNet) (reference).
Arguments
dropout_prob
: the dropout probability in the classifier head. Set tonothing
to disable dropout.inchannels
: the number of input channelsnclasses
: the number of output classesbatchnorm
: set totrue
to include batch normalization after each convolutionbias
: set totrue
to use bias in the convolution layers
Metalhead.inceptionv3
— Functioninceptionv3(; dropout_prob = 0.2, inchannels::Integer = 3, nclasses::Integer = 1000)
Create an Inception-v3 model (reference).
Arguments
dropout_prob
: the dropout probability in the classifier head. Set tonothing
to disable dropout.inchannels
: number of input feature mapsnclasses
: the number of output classes
Metalhead.inceptionv4
— Functioninceptionv4(; dropout_prob = nothing, inchannels::Integer = 3, nclasses::Integer = 1000)
Create an Inceptionv4 model. (reference)
Arguments
dropout_prob
: probability of dropout in classifier head. Set tonothing
to disable dropout.inchannels
: number of input channels.nclasses
: the number of output classes.
Metalhead.inceptionresnetv2
— Functioninceptionresnetv2(; inchannels::Integer = 3, dropout_prob = nothing, nclasses::Integer = 1000)
Creates an InceptionResNetv2 model. (reference)
Arguments
dropout_prob
: probability of dropout in classifier head. Set tonothing
to disable dropout.inchannels
: number of input channels.nclasses
: the number of output classes.
Metalhead.xception
— Functionxception(; dropout_prob = nothing, inchannels::Integer = 3, nclasses::Integer = 1000)
Creates an Xception model. (reference)
Arguments
dropout_prob
: probability of dropout in classifier head. Set tonothing
to disable dropout.inchannels
: number of input channels.nclasses
: the number of output classes.