MethodError: no method matching flatten(::Array{Float32,4})

[paritosh5feb]

I implemented this CNN for Image Segmentation. I am getting the following issue.
`function leNet5()
return Chain(
convBlock(101,202),
MeanPool((5,5)),
convBlock(202,404),
MeanPool((5,5)),
flatten,
Dense(20,40),
Dense(40,80),
Dense(80,101))
end

m = leNet5()
p = params(m)
loss(x, y) = crossentropy(m(x), y)
opt = ADAM(0.001, (0.9, 0.8))
epochs = 10
@epochs epochs Flux.train!(loss, p, train_data, opt, cb = () -> println("training"))`

The Image Data is in the format WHCN.

Error:-

MethodError: no method matching flatten(::Array{Float32,4})
Closest candidates are:
flatten(::AbstractDataFrame, ::Union{Colon, Regex, AbstractString, Signed, Symbol, Unsigned, AbstractArray{T,1} where T, All, Between, Cols, InvertedIndex}) at /home/jupyterlab/.julia/packages/DataFrames/nxjiD/src/abstractdataframe/abstractdataframe.jl:2126

Stacktrace:
[1] macro expansion at /home/jupyterlab/.julia/packages/Zygote/1GXzF/src/compiler/interface2.jl:0 [inlined]
[2] _pullback(::Zygote.Context, ::typeof(flatten), ::Array{Float32,4}) at /home/jupyterlab/.julia/packages/Zygote/1GXzF/src/compiler/interface2.jl:13
[3] applychain at /home/jupyterlab/.julia/packages/Flux/NpkMm/src/layers/basic.jl:30 [inlined]
[4] _pullback(::Zygote.Context, ::typeof(Flux.applychain), ::Tuple{typeof(flatten),Dense{typeof(identity),Array{Float32,2},Array{Float32,1}},Dense{typeof(identity),Array{Float32,2},Array{Float32,1}},Dense{typeof(identity),Array{Float32,2},Array{Float32,1}}}, ::Array{Float32,4}) at /home/jupyterlab/.julia/packages/Zygote/1GXzF/src/compiler/interface2.jl:0
[5] applychain at /home/jupyterlab/.julia/packages/Flux/NpkMm/src/layers/basic.jl:30 [inlined]
[6] _pullback(::Zygote.Context, ::typeof(Flux.applychain), ::Tuple{MeanPool{2,4},typeof(flatten),Dense{typeof(identity),Array{Float32,2},Array{Float32,1}},Dense{typeof(identity),Array{Float32,2},Array{Float32,1}},Dense{typeof(identity),Array{Float32,2},Array{Float32,1}}}, ::Array{Float32,4}) at /home/jupyterlab/.julia/packages/Zygote/1GXzF/src/compiler/interface2.jl:0
[7] applychain at /home/jupyterlab/.julia/packages/Flux/NpkMm/src/layers/basic.jl:30 [inlined]
[8] _pullback(::Zygote.Context, ::typeof(Flux.applychain), ::Tuple{Chain{Tuple{Conv{2,2,typeof(identity),Array{Float32,4},Array{Float32,1}},getfield(Main, Symbol("##101#102"))}},MeanPool{2,4},typeof(flatten),Dense{typeof(identity),Array{Float32,2},Array{Float32,1}},Dense{typeof(identity),Array{Float32,2},Array{Float32,1}},Dense{typeof(identity),Array{Float32,2},Array{Float32,1}}}, ::Array{Float32,4}) at /home/jupyterlab/.julia/packages/Zygote/1GXzF/src/compiler/interface2.jl:0
[9] applychain at /home/jupyterlab/.julia/packages/Flux/NpkMm/src/layers/basic.jl:30 [inlined]
[10] _pullback(::Zygote.Context, ::typeof(Flux.applychain), ::Tuple{MeanPool{2,4},Chain{Tuple{Conv{2,2,typeof(identity),Array{Float32,4},Array{Float32,1}},getfield(Main, Symbol("##101#102"))}},MeanPool{2,4},typeof(flatten),Dense{typeof(identity),Array{Float32,2},Array{Float32,1}},Dense{typeof(identity),Array{Float32,2},Array{Float32,1}},Dense{typeof(identity),Array{Float32,2},Array{Float32,1}}}, ::Array{Float32,4}) at /home/jupyterlab/.julia/packages/Zygote/1GXzF/src/compiler/interface2.jl:0
[11] applychain at /home/jupyterlab/.julia/packages/Flux/NpkMm/src/layers/basic.jl:30 [inlined]
[12] _pullback(::Zygote.Context, ::typeof(Flux.applychain), ::Tuple{Chain{Tuple{Conv{2,2,typeof(identity),Array{Float32,4},Array{Float32,1}},getfield(Main, Symbol("##101#102"))}},MeanPool{2,4},Chain{Tuple{Conv{2,2,typeof(identity),Array{Float32,4},Array{Float32,1}},getfield(Main, Symbol("##101#102"))}},MeanPool{2,4},typeof(flatten),Dense{typeof(identity),Array{Float32,2},Array{Float32,1}},Dense{typeof(identity),Array{Float32,2},Array{Float32,1}},Dense{typeof(identity),Array{Float32,2},Array{Float32,1}}}, ::Array{Float32,4}) at /home/jupyterlab/.julia/packages/Zygote/1GXzF/src/compiler/interface2.jl:0
[13] Chain at /home/jupyterlab/.julia/packages/Flux/NpkMm/src/layers/basic.jl:32 [inlined]
[14] _pullback(::Zygote.Context, ::Chain{Tuple{Chain{Tuple{Conv{2,2,typeof(identity),Array{Float32,4},Array{Float32,1}},getfield(Main, Symbol("##101#102"))}},MeanPool{2,4},Chain{Tuple{Conv{2,2,typeof(identity),Array{Float32,4},Array{Float32,1}},getfield(Main, Symbol("##101#102"))}},MeanPool{2,4},typeof(flatten),Dense{typeof(identity),Array{Float32,2},Array{Float32,1}},Dense{typeof(identity),Array{Float32,2},Array{Float32,1}},Dense{typeof(identity),Array{Float32,2},Array{Float32,1}}}}, ::Array{Float32,4}) at /home/jupyterlab/.julia/packages/Zygote/1GXzF/src/compiler/interface2.jl:0
[15] loss at ./In[41]:4 [inlined]
[16] _pullback(::Zygote.Context, ::typeof(loss), ::Array{Float32,4}, ::Array{Float32,4}) at /home/jupyterlab/.julia/packages/Zygote/1GXzF/src/compiler/interface2.jl:0
[17] adjoint(::Zygote.Context, ::typeof(Core._apply), ::Function, ::Tuple{Array{Float32,4},Array{Float32,4}}) at /home/jupyterlab/.julia/packages/Zygote/1GXzF/src/lib/lib.jl:168
[18] _pullback(::Zygote.Context, ::getfield(Flux.Optimise, Symbol("##17#25")){typeof(loss)}) at /home/jupyterlab/.julia/packages/ZygoteRules/OjfTt/src/adjoint.jl:57
[19] pullback(::Function, ::Params) at /home/jupyterlab/.julia/packages/Zygote/1GXzF/src/compiler/interface.jl:172
[20] gradient(::Function, ::Params) at /home/jupyterlab/.julia/packages/Zygote/1GXzF/src/compiler/interface.jl:53
[21] macro expansion at /home/jupyterlab/.julia/packages/Flux/NpkMm/src/optimise/train.jl:87 [inlined]
[22] macro expansion at /home/jupyterlab/.julia/packages/Juno/n6wyj/src/progress.jl:134 [inlined]
[23] #train!#12(::getfield(Main, Symbol("##121#125")), ::Function, ::typeof(loss), ::Params, ::Flux.Data.DataLoader, ::ADAM) at /home/jupyterlab/.julia/packages/Flux/NpkMm/src/optimise/train.jl:80
[24] (::getfield(Flux.Optimise, Symbol("#kw##train!")))(::NamedTuple{(:cb,),Tuple{getfield(Main, Symbol("##121#125"))}}, ::typeof(Flux.Optimise.train!), ::Function, ::Params, ::Flux.Data.DataLoader, ::ADAM) at ./none:0
[25] top-level scope at /home/jupyterlab/.julia/packages/Flux/NpkMm/src/optimise/train.jl:120
[26] top-level scope at /home/jupyterlab/.julia/packages/Juno/n6wyj/src/progress.jl:134
[27] top-level scope at In[64]:1

[DhairyaLGandhi]

It would be more helpful to check the documentation for flatten. You also probably only need a reshape in this case. I'm closing this issue for now, but feel free to ask usage questions on the julia lang slack on the #helpdesk channel. Thanks!