Error when displaying tensors in REPL

[baggepinnen]

When running the basic usage example in the readme, the error below is produced.

julia> w = exp(x + z + -y)
Error showing value of type TensorFlow.Tensor:
ERROR: BoundsError: attempt to access 0-element Array{Nullable{Int64},1} at index [1]
 in getindex(::TensorFlow.ShapeInference.TensorShape, ::Int64) at /home/fredrikb/.julia/v0.5/TensorFlow/src/shape_inference.jl:29
 in get_shape(::TensorFlow.Tensor) at /home/fredrikb/.julia/v0.5/TensorFlow/src/shape_inference.jl:84
 in (::TensorFlow.ShapeInference.##15#16)(::TensorFlow.Operation) at /home/fredrikb/.julia/v0.5/TensorFlow/src/shape_inference.jl:138
 in get_shape(::TensorFlow.Tensor) at /home/fredrikb/.julia/v0.5/TensorFlow/src/shape_inference.jl:84
 in (::TensorFlow.ShapeInference.##13#14)(::TensorFlow.Operation) at /home/fredrikb/.julia/v0.5/TensorFlow/src/shape_inference.jl:132
 in get_shape(::TensorFlow.Tensor) at /home/fredrikb/.julia/v0.5/TensorFlow/src/shape_inference.jl:84
 in show(::IOContext{Base.Terminals.TTYTerminal}, ::TensorFlow.Tensor) at /home/fredrikb/.julia/v0.5/TensorFlow/src/core.jl:709
 in display(::Base.REPL.REPLDisplay{Base.REPL.LineEditREPL}, ::MIME{Symbol("text/plain")}, ::TensorFlow.Tensor) at ./REPL.jl:132
 in display(::Base.REPL.REPLDisplay{Base.REPL.LineEditREPL}, ::TensorFlow.Tensor) at ./REPL.jl:135
 in display(::TensorFlow.Tensor) at ./multimedia.jl:143
 in print_response(::Base.Terminals.TTYTerminal, ::Any, ::Void, ::Bool, ::Bool, ::Void) at ./REPL.jl:154
 in print_response(::Base.REPL.LineEditREPL, ::Any, ::Void, ::Bool, ::Bool) at ./REPL.jl:139
 in (::Base.REPL.##22#23{Bool,Base.REPL.##33#42{Base.REPL.LineEditREPL,Base.REPL.REPLHistoryProvider},Base.REPL.LineEditREPL,Base.LineEdit.Prompt})(::Base.LineEdit.MIState, ::Base.AbstractIOBuffer{Array{UInt8,1}}, ::Bool) at ./REPL.jl:652
 in (::Base.REPL.##38#47)(::Base.LineEdit.MIState, ::Base.REPL.LineEditREPL, ::Vararg{Any,N}) at ./REPL.jl:867
 in (::Base.LineEdit.##13#14{Base.REPL.##38#47,String})(::Base.LineEdit.MIState, ::Base.REPL.LineEditREPL) at ./LineEdit.jl:736
 in prompt!(::Base.Terminals.TTYTerminal, ::Base.LineEdit.ModalInterface, ::Base.LineEdit.MIState) at ./LineEdit.jl:1605
 in run_interface(::Base.Terminals.TTYTerminal, ::Base.LineEdit.ModalInterface) at ./LineEdit.jl:1574
 in run_frontend(::Base.REPL.LineEditREPL, ::Base.REPL.REPLBackendRef) at ./REPL.jl:903
 in run_repl(::Base.REPL.LineEditREPL, ::Base.##930#931) at ./REPL.jl:188
 in _start() at ./client.jl:360

[baggepinnen]

This is still an issue for this case

input_data = placeholder(Int32, shape=[20, 20])
embedding = get_variable("embedding", [10000, 200], Float32)
inputs = gather(embedding, input_data)
inputs = TensorFlow.split(1, 20, inputs)

20-element Array{TensorFlow.Tensor,1}:
Error showing value of type Array{TensorFlow.Tensor,1}:
ERROR: BoundsError: attempt to access 1-element Array{TensorFlow.ShapeInference.TensorShape,1} at index [2]
 in getindex(::Array{TensorFlow.ShapeInference.TensorShape,1}, ::Int64) at ./array.jl:386
 in get_shape(::TensorFlow.Tensor) at /local/home/fredrikb/.julia/v0.5/TensorFlow/src/shape_inference.jl:81
 in show(::IOContext{Base.AbstractIOBuffer{Array{UInt8,1}}}, ::TensorFlow.Tensor) at /local/home/fredrikb/.julia/v0.5/TensorFlow/src/core.jl:711
 in #sprint#304(::IOContext{Base.Terminals.TTYTerminal}, ::Function, ::Int64, ::Function, ::TensorFlow.Tensor, ::Vararg{TensorFlow.Tensor,N}) at ./strings/io.jl:35
 in (::Base.#kw##sprint)(::Array{Any,1}, ::Base.#sprint, ::Int64, ::Function, ::TensorFlow.Tensor, ::Vararg{TensorFlow.Tensor,N}) at ./<missing>:0
 in alignment(::IOContext{Base.Terminals.TTYTerminal}, ::Array{TensorFlow.Tensor,1}, ::Array{Int64,1}, ::Array{Int64,1}, ::Int64, ::Int64, ::Int64) at ./show.jl:1278
 in print_matrix(::IOContext{Base.Terminals.TTYTerminal}, ::Array{TensorFlow.Tensor,1}, ::String, ::String, ::String, ::String, ::String, ::String, ::Int64, ::Int64) at ./show.jl:1406
 in print_matrix(::IOContext{Base.Terminals.TTYTerminal}, ::Array{TensorFlow.Tensor,1}, ::String, ::String, ::String) at ./show.jl:1378
 in #showarray#330(::Bool, ::Function, ::IOContext{Base.Terminals.TTYTerminal}, ::Array{TensorFlow.Tensor,1}, ::Bool) at ./show.jl:1617
 in display(::Base.REPL.REPLDisplay{Base.REPL.LineEditREPL}, ::MIME{Symbol("text/plain")}, ::Array{TensorFlow.Tensor,1}) at ./REPL.jl:132