Chapter 5 - corrections

[entropidelic]

• In Julia, there are a few PPLs being developed, and we will be using two of them, Turing.l!jl and Soss.jl. We will be focusing on some examples to explain the general approach when using this tools.

  Here we mention that we are going to use Soss.jl, but we ended up only using Turing.jl. This should be fixed unless we add Soss.jl to the chapter.

• We should mention the libraries we are going to use beside Turing.jl and import them in the code. These libraries are:

  • Distributions: For everything related to the sampling from common probability distributions.
  • StatsPlots: For plotting distributions.
  • Turing (It was mentioned but not imported).
    So the block:

  using Distributions
  using StatsPlots
  using Turing
  

  should be in some place at the beginning of the chapter.

• The plotting code should be more consistent; for setting the labels we sometimes do that inside the plot() function with the xlabel and ylabel keyword arguments and sometimes we do it outside the plotting functions with the xlabel!() and ylabel!() calls. I think we should do it all inside. This should be done for all the plotting code of the chapter.

• Change all this code:

  begin
       p_summary = chain[:p]
       plot(p_summary, seriestype = :histogram, normed=true, legend=false, size=(500, 250), color="purple", alpha=0.7)
       title!("Posterior distribution of p after getting tails")
       ylabel!("Probability")
       xlabel!("p")
  end
  

  to

  histogram(chain[:p], xlabel="p", ylabel="Probability", title="Posterior distribution of p after getting tails")
  

• Change this code:

  begin
      plots = histogram.(samples, normalized=true, legend=false, bins=10);
      p_ = plot(plots[1], plots[2], plots[3], plots[4], plots[5], plots[6], plots[7],plots[8], plots[9], layout = 9 , title = ["$i outcomes" for j in 1:1, i in 2:10], titleloc = :right, titlefont = font(8))
  end
  

  to

  plots = [histogram(samples[i], normalized=true, legend=false, bins=10, title="$(i+1) outcomes", titlefont = font(8)) for i in 1:9];
  plot(plots...)
  

• Change this code:

  begin
      plots_ = histogram.(samples_beta_prior, normalized=true, legend=false, bins=10, color="red", alpha=0.6);
      p__ = plot(plots_[1], plots_[2], plots_[3], plots_[4], plots_[5], plots_[6], plots_[7], plots_[8], plots_[9], layout = 9, title = ["$i outcomes" for j in 1:1, i in 2:10], titleloc = :right, titlefont = font(8))
  end
  

  to

   plots = [histogram(samples_beta_prior[i], normalized=true, legend=false, bins=10, title="$(i+1) outcomes", titlefont = font(10), color="red", alpha=0.6) for i in 1:9];