testondata.jl split into computing function and drowing finction

README.md

@@ -219,11 +219,25 @@ All result files are saved in /res directory. To plot a graph run /res/plotcompt
 
 For the computional example on data use the following.
 
-The script `gandata.jl` generates `t = 100000000` realisations of `n = 4` dimensional data form the `t`-multivariate distribution with `\nu = 14` degrees of freedom, and theoretical 
+The script `gandata.jl` generates `t = 150000000` realisations of `n = 4` dimensional data form the `t`-multivariate distribution with `\nu = 14` degrees of freedom, and theoretical 
 super-diagonal elements of those cumulants. Rasults are saved in `data/datafortests.jld`
 
 The script `testondata.jl` computes cumulant tensors of order `m = 1 - 6` for `data/datafortests.jld`, results are saved in `data/cumulants.jld`.
 
+To read `cumulants.jld` please run 
+
+```julia 
+julia> using JLD
+
+julia> using SymmetricTensors
+
+julia> load("cumulants.jld")
+
+```
+
+To plot super-diagonal elements of those cumulants and their theoretical values from t-student dostrobution pleas run `plotsuperdiag.jl`
+
+
 # Citing this work
 
 

test/data/plotsuperdiag.jl

@@ -0,0 +1,48 @@
+#!/usr/bin/env julia
+
+using JLD
+using SymmetricTensors
+using PyCall
+@pyimport matplotlib as mpl
+using PyPlot
+mpl.rc("text", usetex=true)
+mpl.rc("font", family="serif", size = 8)
+
+
+"""
+  pltdiag()
+
+Plots a chart of superdiagonal elements of cumulants of and its theoretical values
+"""
+function pltdiag()
+  tdiag = try load("datafortests.jld")["theoretical diag"]
+  catch
+    println("please run test/gandata.jl and test/testondata.jl")
+  return ()
+  end
+  cum = try load("cumulants.jld")["cumulants"]
+  catch
+    println("please run test/testondata.jl")
+    return ()
+  end
+  n = cum[1].dats
+  fig, ax = subplots(figsize = (3., 2.3))
+  col = ["cyan", "brown", "green", "red", "blue", "black"]
+  for order in (2,4,5,6)
+    c = cum[order]
+    ax[:plot](diag(c), "o", color = col[order], label = "$order cumulant", markersize=3)
+    ax[:plot]([fill(tdiag[order], n)...], "--", color = col[order], label = "theoretical")
+  end
+  PyPlot.ylabel("superdiagonal elements", labelpad = -1)
+  PyPlot.xlabel("superdiagonal number", labelpad = -3)
+  ax[:legend](fontsize = 4.5, loc = 5)
+  fig[:savefig]("diagcumels.eps")
+end
+
+
+
+function main()
+  pltdiag()
+end
+
+main()

test/gendata.jl

@@ -10,7 +10,7 @@ Returns Matrix{Float64} - t realisations from t-student multivatiate distributio
 with nu degress of freedom
 """
 
-function gendat(nu::Int, t::Int = 100000000)
+function gendat(nu::Int, t::Int = 150000000)
   cm = [[1. 0.7 0.7 0.7];[0.7 1. 0.7 0.7]; [0.7 0.7 1. 0.7]; [0.7 0.7 0.7 1]]
   p = MvTDist(nu, [0., 0., 0., 0.],cm)
   return transpose(rand(p, t))

test/testondata.jl

@@ -2,7 +2,7 @@
 
 using JLD
 using Cumulants
-addprocs()
+addprocs(4)
 @everywhere using Cumulants
 
 function main()
@@ -12,7 +12,7 @@ function main()
     return ()
   end
   c = cumulants(d["data"], 6)
-  save("data/cumulants.jld", Dict("cumulant" => c))
+  save("data/cumulants.jld", Dict("cumulants" => c))
 end
 
 main()