Error when using ScikitLearn.jl

[yusri-dh]

I have an error when import ScikitLearn model.

Here is my code:

begin
	import Pkg
	Pkg.activate(mktempdir())
	Pkg.add("ScikitLearn")
end

using ScikitLearn

@sk_import linear_model: LogisticRegression

when I performed last line of code above, an error occured:

Failed to show value:

PyError ($(Expr(:escape, :(ccall(#= /home/yusri/.julia/packages/PyCall/zqDXB/src/pyfncall.jl:43 =# @pysym(:PyObject_Call), PyPtr, (PyPtr, PyPtr, PyPtr), o, pyargsptr, kw))))) <class 'TypeError'>

TypeError("'property' object is not callable")

   pyerr_check@exception.jl:60[inlined]
   pyerr_check@exception.jl:64[inlined]
   _handle_error(::String)@exception.jl:81
   macro expansion@exception.jl:95[inlined]
   #110@pyfncall.jl:43[inlined]
   disable_sigint@c.jl:446[inlined]
   __pycall!@pyfncall.jl:42[inlined]
   _pycall!(::PyCall.PyObject, ::PyCall.PyObject, ::Tuple{}, ::Int64, ::Ptr{Nothing})@pyfncall.jl:29
   _pycall!@pyfncall.jl:11[inlined]
   #pycall#115@pyfncall.jl:80[inlined]
   pycall@pyfncall.jl:80[inlined]
   showable(::MIME{Symbol("text/html")}, ::PyCall.PyObject)@PyCall.jl:894
   #invokelatest#1@essentials.jl:710[inlined]
   invokelatest@essentials.jl:709[inlined]
   #17@PlutoRunner.jl:307[inlined]
   iterate@iterators.jl:431[inlined]
   first@abstractarray.jl:341[inlined]
   |>(::Base.Iterators.Filter{Main.PlutoRunner.var"#17#19"{PyCall.PyObject},Array{MIME,1}}, ::typeof(first))@operators.jl:834
   #show_richest#16(::Bool, ::typeof(Main.PlutoRunner.show_richest), ::IOContext{Base.GenericIOBuffer{Array{UInt8,1}}}, ::Any)@PlutoRunner.jl:307
   show_richest(::IOContext{Base.GenericIOBuffer{Array{UInt8,1}}}, ::Any)@PlutoRunner.jl:307
   #sprint_withreturned#15(::IOContext{Base.PipeEndpoint}, ::Int64, ::typeof(Main.PlutoRunner.sprint_withreturned), ::Function, ::PyCall.PyObject)@PlutoRunner.jl:279
   format_output(::Any)@PlutoRunner.jl:219
   formatted_result_of(::Base.UUID, ::Bool)@PlutoRunner.jl:41
   top-level scope@none:1

The code is working well in my repl.

[Moelf]

try slap on a ; to suppress output for now, this is indeed a bug

[fonsp]

Could you run this notebook and tell us what the outputs/errors were?

begin
	import Pkg
	Pkg.activate(mktempdir())
	Pkg.add("ScikitLearn")
end

using ScikitLearn

x = begin
	@sk_import linear_model: LogisticRegression
end;

Base.invokelatest(showable, MIME"text/plain"(), x)

Base.invokelatest(showable, MIME"text/html"(), x)

Base.invokelatest(showable, MIME"text/latex"(), x)

Base.invokelatest(showable, MIME"application/vnd.pluto.tree+xml"(), x)

Base.invokelatest(showable, MIME"image/png"(), x)

[yusri-dh]

@fonsp I run your code in Pluto.jl:

Base.invokelatest(showable, MIME"text/plain"(), x)
>true

Base.invokelatest(showable, MIME"text/html"(), x)
>PyError ($(Expr(:escape, :(ccall(#= /home/yusri/.julia/packages/PyCall/zqDXB/src/pyfncall.jl:43 =# @pysym(:PyObject_Call), PyPtr, (PyPtr, PyPtr, PyPtr), o, pyargsptr, kw))))) <class 'TypeError'>

TypeError("'property' object is not callable")

    pyerr_check@exception.jl:60[inlined]
    pyerr_check@exception.jl:64[inlined]
    _handle_error(::String)@exception.jl:81
    macro expansion@exception.jl:95[inlined]
    #110@pyfncall.jl:43[inlined]
    disable_sigint@c.jl:446[inlined]
    __pycall!@pyfncall.jl:42[inlined]
    _pycall!(::PyCall.PyObject, ::PyCall.PyObject, ::Tuple{}, ::Int64, ::Ptr{Nothing})@pyfncall.jl:29
    _pycall!@pyfncall.jl:11[inlined]
    #pycall#115@pyfncall.jl:80[inlined]
    pycall@pyfncall.jl:80[inlined]
    showable(::MIME{Symbol("text/html")}, ::PyCall.PyObject)@PyCall.jl:894
    #invokelatest#1(::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}, ::typeof(Base.invokelatest), ::Any, ::Any, ::Vararg{Any,N} where N)@essentials.jl:710
    invokelatest(::Any, ::Any, ::Vararg{Any,N} where N)@essentials.jl:709
    top-level scope@Local: 1

Base.invokelatest(showable, MIME"text/latex"(), x)
>false

Base.invokelatest(showable, MIME"application/vnd.pluto.tree+xml"(), x)
>false

Base.invokelatest(showable, MIME"image/png"(), x)
>false

[fonsp]

You should open a bug report with ScikitLearn.jl - this is unrelated to Pluto.

Here's the MWE of the bug:

using ScikitLearn

x = begin
	@sk_import linear_model: LogisticRegression
end

showable(MIME"text/html"(), x)

[cstjean]

try slap on a ; to suppress output for now, this is indeed a bug

@yusri-dh Does that fix the issue?