Bug of Running mnistexample.jl

[tangzhenyu]

ERROR: LoadError: UndefVarError: axpy! not defined
in #calculate_weight_gradient!#17(::ASCIIString, ::Any, ::Boltzmann.RBM{Distributions.Bernoulli,Distributions.Bernoulli}, ::Array{Float64,2}, ::Array{Float64,2}, ::Array{Float64,2}, ::Array{Float64,2}, ::Float64) at /home/tangzhenyu/.julia/v0.5/Boltzmann/src/Training.jl:30
[inlined code] from ./boot.jl:307
in #fit_batch!#19(::Bool, ::Float64, ::Int64, ::ASCIIString, ::Float64, ::ASCIIString, ::Any, ::Boltzmann.RBM{Distributions.Bernoulli,Distributions.Bernoulli}, ::Array{Float64,2}) at /home/tangzhenyu/.julia/v0.5/Boltzmann/src/Training.jl:130
[inlined code] from ./boot.jl:307
in #fit#20(::Bool, ::Float64, ::Int64, ::Int64, ::Int64, ::ASCIIString, ::Float64, ::Array{Any,1}, ::Int64, ::Bool, ::ASCIIString, ::Int64, ::Void, ::StatsBase.#fit, ::Boltzmann.RBM{Distributions.Bernoulli,Distributions.Bernoulli}, ::Array{Float64,2}) at /home/tangzhenyu/.julia/v0.5/Boltzmann/src/Training.jl:264
[inlined code] from ./boot.jl:307
in run_mnist() at /home/tangzhenyu/.julia/v0.5/Boltzmann/examples/mnistexample.jl:32
in include(::ASCIIString) at ./boot.jl:240
in include_from_node1(::UTF8String) at ./loading.jl:417
in process_options(::Base.JLOptions) at ./client.jl:262
in _start() at ./client.jl:318
while loading /home/tangzhenyu/.julia/v0.5/Boltzmann/examples/mnistexample.jl, in expression starting on line 44

[eric-tramel]

Hi @tangzhenyu , thanks for letting us know about this issue, sorry for the delayed response. Can you let us know a bit about your environment?

Specifically, which release/build of Julian are you using and on what system? Did you build Julia from source or did you grab a binary? It seems that there is some issue here with Julia seeing the BLAS libraries. Perhaps your current build of Julia is missing s BLAS distribution?

[tangzhenyu]

@eric-tramel ,very thanks for your reminder.Actually,as you are saying,it was lacking of BLAS libraries.After installed this library,the code works.Thanks a lot again.