On Tuesday, November 22, 2016 at 5:53:14 PM UTC-2, Kevin Liu wrote: > On Friday, November 18, 2016 at 4:07:44 PM UTC-2, Kevin Liu wrote: > > Have a look please https://github.com/hpoit/MLN.jl/tree/master/BN > > > > On Friday, November 18, 2016 at 11:48:58 AM UTC-2, Yichao Yu wrote:On Thu, > > Nov 17, 2016 at 2:39 PM, Kevin Liu <kvt...@gmail.com> wrote: > > > > > Right, I need the instance of Factor > > > > > > > > Then use the instance of Factor. > > > > > > > > > > > > > > On Thursday, November 17, 2016 at 5:33:05 PM UTC-2, Yichao Yu wrote: > > > > >> > > > > >> On Thu, Nov 17, 2016 at 2:27 PM, Kevin Liu <kvt...@gmail.com> wrote: > > > > >> > I replaced Factor[:FactorMargin]() with Factor.FactorMargin() back > > > > >> > again. > > > > >> > > > > > >> > Still, for FactorOperations.jl on Atom, I get {UndefVarError: Factor > > >> > not > > > > >> > defined} at the end of each block. > > > > >> > > > > > >> > Factor is defined on Factor.jl, and that file evaluates fine. > > > > >> > > > > >> AFAICT the `Factor`is a type so `Factor.FactorMargin` is definitely > > >> wrong. > > > > >> > > > > >> > > > > > >> > The main file, BN.jl, includes Factor.jl and FactorOperations.jl and > > > > >> > exports > > > > >> > Factor, and also evaluates fine. > > > > >> > > > > > >> > On Wednesday, November 16, 2016 at 11:25:24 PM UTC-2, Yichao Yu wrote: > > > > >> >> > > > > >> >> On Wed, Nov 16, 2016 at 7:24 PM, Kevin Liu <kvt...@gmail.com> wrote: > > > > >> >> > Hi Yichao! > > > > >> >> > > > > >> >> In general there's nothing from the code you posted that shows what > > > > >> >> you want to do. > > > > >> >> > > > > >> >> > > > > > >> >> > Here is the function from FactorOperations.jl > > > > >> >> > > > > > >> >> > function FactorDropMargin(A::Factor, Remove_var::Vector{String}) > > > > >> >> > Remove_dims = indexin(Remove_var, A.var) > > > > >> >> > if any(Remove_dims==0) > > > > >> >> > error("Wrong variable!") > > > > >> >> > end > > > > >> >> > > > > > >> >> > Remain_var = symdiff(A.var, Remove_var) > > > > >> >> > Remain_dims = indexin(Remain_var, A.var) > > > > >> >> > > > > > >> >> > Factor[:FactorMargin](A, Remove_var, Remain_var, Remove_dims, > > > > >> >> > Remain_dims) # line 85 > > > > >> >> > > > > >> >> Unless you overloaded getindex on this type (which you should include) > > > > >> >> you are construction a Vector of `Factor` from a symbol and then > > > > >> >> calling it. It's impossible to tell what you actually want to do. > > > > >> >> > > > > >> >> And as I previously mentioned, unless you are using PyCall, the issue > > > > >> >> you linked is totally unrelated to this. > > > > >> >> > > > > >> >> > end > > > > >> >> > > > > > >> >> > runtests.jl: > > > > >> >> > > > > > >> >> > @testset "Multiply and marginalize factor" begin > > > > >> >> > > > > > >> >> > A=Factor(["a", "b"],[3, 2],[0.5, 0.1, 0.3, 0.8, 0, 0.9]) > > > > >> >> > B=Factor(["b", "c"],[2, 2],[0.5, 0.1, 0.7, 0.2]) > > > > >> >> > C = FactorProduct(A, B) > > > > >> >> > FactorDropMargin(C, ["c"]) # line 19 > > > > >> >> > FactorKeepMargin(C, ["b", "a"]) > > > > >> >> > FactorPermute(ans, [2, 1]) > > > > >> >> > FactorKeepMargin(C, ["a", "b"]) > > > > >> >> > > > > > >> >> > end > > > > >> >> > > > > > >> >> > what I got on the REPL: > > > > >> >> > > > > > >> >> > julia> Pkg.test("BN") > > > > >> >> > > > > > >> >> > INFO: Testing BN > > > > >> >> > > > > > >> >> > Test Summary: | > > > > >> >> > > > > > >> >> > Define, permute factor, and call (var, card, val) | No tests > > > > >> >> > > > > > >> >> > Multiply and marginalize factor: Error During Test > > > > >> >> > > > > > >> >> > Got an exception of type ErrorException outside of a @test > > > > >> >> > > > > > >> >> > type DataType has no field FactorMargin > > > > >> >> > > > > > >> >> > in FactorDropMargin(::BN.Factor, ::Array{String,1}) at > > > > >> >> > /Users/Corvus/.julia/v0.5/BN/src/FactorOperations.jl:85 > > > > >> >> > > > > > >> >> > in macro expansion; at > > > > >> >> > /Users/Corvus/.julia/v0.5/BN/test/runtests.jl:19 > > > > >> >> > [inlined] > > > > >> >> > > > > > >> >> > in macro expansion; at ./test.jl:672 [inlined] > > > > >> >> > > > > > >> >> > in anonymous at ./<missing>:? > > > > >> >> > > > > > >> >> > in include_from_node1(::String) at ./loading.jl:488 > > > > >> >> > > > > > >> >> > in include_from_node1(::String) at > > > > >> >> > > > > > >> >> > > > > > >> >> > /Applications/Julia-0.5.app/Contents/Resources/julia/lib/julia/sys.dylib:? > > > > >> >> > > > > > >> >> > in process_options(::Base.JLOptions) at ./client.jl:262 > > > > >> >> > > > > > >> >> > in _start() at ./client.jl:318 > > > > >> >> > > > > > >> >> > in _start() at > > > > >> >> > > > > > >> >> > > > > > >> >> > /Applications/Julia-0.5.app/Contents/Resources/julia/lib/julia/sys.dylib:? > > > > >> >> > > > > > >> >> > Test Summary: | Error Total > > > > >> >> > > > > > >> >> > Multiply and marginalize factor | 1 1 > > > > >> >> > > > > > >> >> > > > > > >> >> > On Wednesday, November 16, 2016 at 10:02:48 PM UTC-2, Yichao Yu > > > > >> >> > wrote: > > > > >> >> >> > > > > >> >> >> On Wed, Nov 16, 2016 at 6:50 PM, Kevin Liu <kvt...@gmail.com> > > >> >> >> wrote: > > > > >> >> >> > From this issue https://github.com/JuliaPy/PyPlot.jl/issues/157 I > > > > >> >> >> > understand > > > > >> >> >> > > > > >> >> >> ^^ This is irrelevant unless you are using PyCall > > > > >> >> >> > > > > >> >> >> > > > > > >> >> >> > `Factor[:FactorMargin](A, Remove_var, Remain_var, Remove_dims, > > > > >> >> >> > Remain_dims)` > > > > >> >> >> > (line 85 of FactorOperations.jl) should pass, as it does on Atom, > > > > >> >> >> > but > > > > >> >> >> > not on > > > > >> >> >> > the REPL, which throws > > > > >> >> >> > > > > > >> >> >> > Got an exception of type ErrorException outside of a @test > > > > >> >> >> > > > > > >> >> >> > type DataType has no field FactorMargin > > > > >> >> >> > > > > > >> >> >> > in FactorDropMargin(::BN.Factor, ::Array{String,1}) at > > > > >> >> >> > /Users/Corvus/.julia/v0.5/BN/src/FactorOperations.jl:85 > > > > >> >> >> > > > > >> >> >> Impossible to tell without code. > > > > >> >> >> > > > > >> >> >> > > > > > >> >> >> > > > > > >> >> >> > Help, please. > > Yichao and DPSanders, I have already used instances of Factor on runtests.jl, > instances A, B, and C > > A=Factor(["a", "b"],[3, 2],[0.5, 0.1, 0.3, 0.8, 0, 0.9]) > B=Factor(["b", "c"],[2, 2],[0.5, 0.1, 0.7, 0.2]) > C = FactorProduct(A, B) > FactorDropMargin(C, ["c"]) > > Do you recommend I make any of the functions in FactorOperations.jl into > inner constructors of `type Factor` in Factor.jl?
What the tests should look like julia> A=Factor(["a", "b"],[3, 2],[0.5, 0.1, 0.3, 0.8, 0, 0.9]); julia> B=Factor(["b", "c"],[2, 2],[0.5, 0.1, 0.7, 0.2]); julia> C = FactorProduct(A, B) Factor(["a", "b", "c"],[3, 2, 2],[0.25, 0.05, 0.15, 0.08, 0.0, 0.09, 0.35, 0.07, 0.21, 0.16, 0.0, 0.18]) julia> FactorDropMargin(C, ["c"]) Factor(["a", "b"],[3, 2],[0.6, 0.12, 0.36, 0.24, 0.0, 0.27]) julia> FactorKeepMargin(C, ["b", "a"]) Factor(["b", "a"],[2, 3],[0.6, 0.24, 0.12, 0.0, 0.36, 0.27]) julia> FactorPermute(ans, [2, 1]) Factor(["a", "b"],[3, 2],[0.6, 0.12, 0.36, 0.24, 0.0, 0.27]) julia> FactorKeepMargin(C, ["a", "b"]) Factor(["a", "b"],[3, 2],[0.6, 0.12, 0.36, 0.24, 0.0, 0.27])
