ArrayDataProvider returns TypeError

[machiningcentre]

ArrayDataProvider returns TypeError when using Images.jl.

julia> using Images
julia> using MXNet
julia> X = rand(10,10);
julia> p = mx.ArrayDataProvider(X)
ERROR: TypeError: #ArrayDataProvider#6688: in new, expected Array{Array{Float32,N},1}, got Array{Array{Float32,N},1}
 in #ArrayDataProvider#6688(::Int64, ::Bool, ::Int64, ::Int64, ::Type{T}, ::Array{Float64,2}, ::Array{Any,1}) at ~/.julia/v0.5/MXNet/src/io.jl:351
 in (::Core.#kw#Type)(::Array{Any,1}, ::Type{MXNet.mx.ArrayDataProvider}, ::Array{Float64,2}, ::Array{Any,1}) at ./<missing>:0
 in MXNet.mx.ArrayDataProvider(::Array{Float64,2}) at ~/.julia/v0.5/MXNet/src/io.jl:277

There is no problem without Images.jl.

julia> using MXNet
julia> X = rand(10,10);
julia> p = mx.ArrayDataProvider(X)
MXNet.mx.ArrayDataProvider(Array{Float32,N}[
Float32[0.415866 0.112956 … 0837056 0.398794; 0.590402 0.334468 … 0515418 0.200105; … ; 0.3765410.337045 … 0.677437 0.31439; 0.761557 0.601104 … 0.7393450.0729888]],Symbol[:data],Array{Float32,N}[],Symbol[],10,10,false,0.0f0,0.0f0,MXNet.mx.NDArray[mx.NDArray{Float32}(10,10)],MXNet.mx.NDArray[])

Julia version and the package status are as follows:

julia> versioninfo()                                        
Julia Version 0.5.0                                         
Commit 3c9d753 (2016-09-19 18:14 UTC)                       
Platform Info:                                              
  System: Linux (x86_64-pc-linux-gnu)                       
  CPU: Intel(R) Xeon(R) CPU E5-1650 v3 @ 3.50GHz            
  WORD_SIZE: 64                                             
  BLAS: libopenblas (USE64BITINT DYNAMIC_ARCH NO_AFFINITY Haswell)
  LAPACK: libopenblas64_                                    
  LIBM: libopenlibm                                         
  LLVM: libLLVM-3.7.1 (ORCJIT, haswell)                     
                                                            
julia> Pkg.status("MXNet")                                  
 - MXNet                         0.2.1                      
                                                            
julia> Pkg.status("Images")                                 
 - Images                        0.7.0                      

[pluskid]

I can confirm this error. However, I do not understand the error message or why such an error is thrown out. Maybe we should open an issue in the Julia repo or Images.jl repo to see if anyone would know what is the issue here.

[xafilox]

Hi. I'm having the same problem. Any news about it? Were you able to solve it?

[vchuravy]

What version of Julia are you using? What is you Pkg.status()? I noticed this as well and it might be related to JuliaLang/julia#18465

cc: @timholy

[xafilox]

Hi. My version and packages are the following:

Julia Version 0.5.0
Commit 3c9d753 (2016-09-19 18:14 UTC)
Platform Info:
System: Linux (x86_64-linux-gnu)
CPU: Intel(R) Core(TM) i7-6950X CPU @ 3.00GHz
WORD_SIZE: 64
BLAS: libopenblas (NO_LAPACKE DYNAMIC_ARCH NO_AFFINITY Haswell)
LAPACK: liblapack.so.3
LIBM: libopenlibm
LLVM: libLLVM-3.7.1 (ORCJIT, broadwell)

17 required packages:

• ArrayFire 0.0.4
• Feather 0.2.5
• Gadfly 0.6.0
• GraphViz 0.1.0
• IJulia 1.4.1
• ImageMagick 0.2.3
• JLD 0.6.10
• MXNet 0.2.1
• MultivariateStats 0.3.1
• Plots 0.10.3
• Query 0.3.2
• RDatasets 0.2.0
• Requests 0.4.1
• StructuredQueries 0.0.4
• TensorFlow 0.5.1
• XGBoost 0.2.0
• ZipFile 0.3.0
  83 additional packages:
• AxisAlgorithms 0.1.6
• BinDeps 0.4.7
• Blosc 0.2.0
• BufferedStreams 0.3.2
• Calculus 0.2.2
• CategoricalArrays 0.1.3
• Codecs 0.3.0
• ColorTypes 0.4.0
• Colors 0.7.3
• Compat 0.22.0
• Compose 0.4.5
• Conda 0.5.3
• Contour 0.2.0
• DataArrays 0.3.12
• DataFrames 0.9.0
• DataStreams 0.1.3
• DataStructures 0.5.3
• DiffBase 0.1.0
• Distances 0.4.1
• Distributions 0.12.2
• DocStringExtensions 0.3.2
• Documenter 0.9.2
• DualNumbers 0.3.0
• FileIO 0.3.1
• FixedPointNumbers 0.3.6
• FixedSizeArrays 0.2.5
• FlatBuffers 0.2.0
• Formatting 0.2.1
• ForwardDiff 0.4.1
• FunctionWrappers 0.0.1
• GZip 0.3.0
• Graphics 0.2.0
• HDF5 0.8.0
• Hexagons 0.0.4
• Hiccup 0.1.1
• HttpCommon 0.2.7
• HttpParser 0.2.0
• ImageCore 0.2.1
• Interpolations 0.3.8
• Iterators 0.3.0
• JSON 0.8.3
• Juno 0.2.7
• KernelDensity 0.3.2
• LegacyStrings 0.2.1
• Libz 0.2.4
• LineSearches 0.1.5
• Loess 0.1.0
• MNIST 0.0.2
• MacroTools 0.3.6
• MappedArrays 0.0.7
• MbedTLS 0.4.5
• Measures 0.0.3
• Media 0.2.6
• NaNMath 0.2.4
• NamedTuples 1.0.0
• Nettle 0.3.0
• NullableArrays 0.1.0
• OffsetArrays 0.2.14
• Optim 0.7.8
• PDMats 0.5.6
• PlotThemes 0.1.1
• PlotUtils 0.3.0
• PositiveFactorizations 0.0.4
• ProtoBuf 0.4.0
• PyCall 1.11.1
• QuadGK 0.1.2
• RData 0.0.4
• Ratios 0.0.4
• RecipesBase 0.1.0
• Reexport 0.0.3
• Requires 0.3.0
• Rmath 0.1.6
• SHA 0.3.2
• ShowItLikeYouBuildIt 0.0.1
• Showoff 0.0.7
• SortingAlgorithms 0.1.1
• SpecialFunctions 0.1.1
• StatsBase 0.13.1
• StatsFuns 0.4.0
• URIParser 0.1.8
• WeakRefStrings 0.2.0
• WoodburyMatrices 0.2.2
• ZMQ 0.4.2

[vchuravy]

I can confirm that this happens on 0.5.1. as well so it is unrelated to the issue I noted above.