ConvTranspose(filter, in => out, σ=identity; stride=1, pad=0, dilation=1, [bias, init])

Standard convolutional transpose layer. filter is a tuple of integers specifying the size of the convolutional kernel, while in and out specify the number of input and output channels.

Note that pad=SamePad() here tries to ensure size(output,d) == size(x,d) * stride.

Parameters are controlled by additional keywords, with defaults init=glorot_uniform and bias=true.

See also Conv for more detailed description of keywords.

Examples

			julia> xs = rand32(100, 100, 3, 50);  # a batch of 50 RGB images

julia> layer = ConvTranspose((5,5), 3 => 7, relu)
ConvTranspose((5, 5), 3 => 7, relu)  # 532 parameters

julia> layer(xs) |> size
(104, 104, 7, 50)

julia> ConvTranspose((5,5), 3 => 7, stride=2)(xs) |> size
(203, 203, 7, 50)

julia> ConvTranspose((5,5), 3 => 7, stride=3, pad=SamePad())(xs) |> size
(300, 300, 7, 50)

			ConvTranspose(weight::AbstractArray, [bias, activation; stride, pad, dilation, groups])

Constructs a ConvTranspose layer with the given weight and bias. Accepts the same keywords and has the same defaults as [ ConvTranspose(k::NTuple{N,Integer}, ch::Pair{<:Integer,<:Integer}, σ; ...)]( ConvTranspose).

Examples

			julia> weight = rand(3, 4, 5);

julia> bias = zeros(4);

julia> layer = ConvTranspose(weight, bias, sigmoid)
ConvTranspose((3,), 5 => 4, σ)  # 64 parameters

julia> layer(randn(100, 5, 64)) |> size  # transposed convolution will increase the dimension size (upsampling)
(102, 4, 64)

julia> Flux.params(layer) |> length
2