I watched the juliacon video on TensorFlow.jl and found it quite cool. So I tried to reproduce the examples.
In the video the visualize function is used… If I try to do the same thing on my machine I get an error however:
could not spawn `tensorboard --logdir=/tmp/tmpdXvsXU --port=6006`: no such file or directory (ENOENT)
Stacktrace:
[1] _jl_spawn(::String, ::Array{String,1}, ::Ptr{Void}, ::Base.Process, ::Base.DevNullStream, ::Base.PipeEndpoint, ::Ptr{Void}) at ./process.jl:360
[2] #373 at ./process.jl:512 [inlined]
[3] setup_stdio(::Base.##373#374{Cmd}, ::Tuple{Base.DevNullStream,Pipe,Base.PipeEndpoint}) at ./process.jl:499
[4] #spawn#372(::Nullable{Base.ProcessChain}, ::Function, ::Cmd, ::Tuple{Base.DevNullStream,Pipe,Base.PipeEndpoint}) at ./process.jl:511
[5] spawn(::Cmd, ::Tuple{Base.DevNullStream,Pipe,Base.PipeEndpoint}) at ./process.jl:506
[6] open(::Cmd, ::String, ::Base.DevNullStream) at ./process.jl:578
[7] get_tensorboard(::Void) at /home/jan/.julia/v0.6/TensorFlow/src/show.jl:148
[8] visualize(::TensorFlow.Graph) at /home/jan/.julia/v0.6/TensorFlow/src/show.jl:179
[9] visualize() at /home/jan/.julia/v0.6/TensorFlow/src/core.jl:250
Is this expected? To install TensorFlow I simply did
Pkg.add("TensorFlow")
Pkg.checkout("TensorFlow")
Pkg.build("TensorFlow")
Do I have to install tensorboard manually or something?