Move images to Makie (#72)

* added .JuliaFormatter.toml

* makie 02_bayes_stats.jl

* makie 03_prob_dist

* format 01_why_Julia

* format everything

* makie conversion

* using stuff multiline

* makie 04_Turing

* remove plots stuff from Project.toml

* animations and makie 05_MCMC

* fix setrange in 12_Turing_tricks

Author: storopoli

.JuliaFormatter.toml

@@ -0,0 +1 @@
+style = "blue"

Project.toml

@@ -6,6 +6,7 @@ CSV = "336ed68f-0bac-5ca0-87d4-7b16caf5d00b"
 CairoMakie = "13f3f980-e62b-5c42-98c6-ff1f3baf88f0"
 CategoricalArrays = "324d7699-5711-5eae-9e2f-1d82baa6b597"
 Chain = "8be319e6-bccf-4806-a6f7-6fae938471bc"
+Colors = "5ae59095-9a9b-59fe-a467-6f913c188581"
 DataFrames = "a93c6f00-e57d-5684-b7b6-d8193f3e46c0"
 DifferentialEquations = "0c46a032-eb83-5123-abaf-570d42b7fbaa"
 Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"
@@ -14,15 +15,12 @@ FillArrays = "1a297f60-69ca-5386-bcde-b61e274b549b"
 ForwardDiff = "f6369f11-7733-5829-9624-2563aa707210"
 Franklin = "713c75ef-9fc9-4b05-94a9-213340da978e"
 HTTP = "cd3eb016-35fb-5094-929b-558a96fad6f3"
-LaTeXStrings = "b964fa9f-0449-5b57-a5c2-d3ea65f4040f"
 LazyArrays = "5078a376-72f3-5289-bfd5-ec5146d43c02"
 Literate = "98b081ad-f1c9-55d3-8b20-4c87d4299306"
 MCMCChains = "c7f686f2-ff18-58e9-bc7b-31028e88f75d"
 NodeJS = "2bd173c7-0d6d-553b-b6af-13a54713934c"
-Plots = "91a5bcdd-55d7-5caf-9e0b-520d859cae80"
 StatsBase = "2913bbd2-ae8a-5f71-8c99-4fb6c76f3a91"
 StatsFuns = "4c63d2b9-4356-54db-8cca-17b64c39e42c"
-StatsPlots = "f3b207a7-027a-5e70-b257-86293d7955fd"
 Turing = "fce5fe82-541a-59a6-adf8-730c64b5f9a0"
 
 [compat]

_literate/01_why_Julia.jl

@@ -76,7 +76,11 @@
 # Here is Julia:
 
 # ```julia
-# using Random, StatsBase, DataFrames, BenchmarkTools, Chain
+# using Random
+# using StatsBase
+# using DataFrames
+# using BenchmarkTools
+# using Chain
 # Random.seed!(123)
 #
 # n = 10_000
@@ -474,12 +478,16 @@
 # to define the action that one type `Pet` takes when it meets another `Pet`:
 
 abstract type Pet end
-struct Dog <: Pet name::String end
-struct Cat <: Pet name::String end
+struct Dog <: Pet
+    name::String
+end
+struct Cat <: Pet
+    name::String
+end
 
 function encounter(a::Pet, b::Pet)
     verb = meets(a, b)
-    println("$(a.name) meets $(b.name) and $verb")
+    return println("$(a.name) meets $(b.name) and $verb")
 end
 
 meets(a::Dog, b::Dog) = "sniffs";
@@ -639,7 +647,7 @@ using BenchmarkTools
 # with a simple column selection. We do this by defining a new method of the `*` function (multiplier function)
 # of `Base` Julia. Additionally we also create a new optimized method of `inner()` for dealing with `OneHotVector`:
 
-import Base:*
+import Base: *
 
 *(A::AbstractMatrix, v::OneHotVector) = A[:, v.ind]
 inner(v::OneHotVector, A, w::OneHotVector) = A[v.ind, w.ind]

_literate/02_bayes_stats.jl

@@ -416,24 +416,23 @@
 # and the 75% percentile of the probability density of $\theta$. In this example, MLE leads to estimated values that are not
 # consistent with the actual probability density of the value of $\theta$.
 
-using Plots, StatsPlots, Distributions, LaTeXStrings
+using CairoMakie
+using Distributions
 
 d = LogNormal(0, 2)
 range_d = 0:0.001:4
 q25 = quantile(d, 0.25)
 q75 = quantile(d, 0.75)
-plot((range_d, pdf.(d, range_d)),
-     leg=false,
-     xlims=(-0.2, 4.2),
-     lw=3,
-     xlabel=L"\theta",
-     ylabel="Density")
-scatter!((mode(d), pdf(d, mode(d))), mc=:green, ms=5)
-plot!(range(q25, stop=q75, length=100),
-      x -> pdf(d, x),
-      lc=false, fc=:blues,
-      fill=true, fa=0.5)
-savefig(joinpath(@OUTPUT, "lognormal.svg")); # hide
+credint = range(q25; stop=q75, length=100)
+f, ax, l = lines(
+    range_d,
+    pdf.(d, range_d);
+    linewidth=3,
+    axis=(; limits=(-0.2, 4.2, nothing, nothing), xlabel=L"\theta", ylabel="Density"),
+)
+scatter!(ax, mode(d), pdf(d, mode(d)); color=:green, markersize=12)
+band!(ax, credint, 0.0, pdf.(d, credint); color=(:steelblue, 0.5))
+save(joinpath(@OUTPUT, "lognormal.svg"), f); # hide
 
 # \fig{lognormal}
 # \center{_**Log-Normal**: Maximum Likelihood Estimate vs Credible Intervals_} \\
@@ -454,19 +453,22 @@ range_d = -2:0.01:14
 sim_d = rand(d, 10_000)
 q25 = quantile(sim_d, 0.25)
 q75 = quantile(sim_d, 0.75)
-plot((range_d, pdf.(d, range_d)),
-     leg=false,
-     xlims=(-2, 14),
-     xticks=[0, 5, 10],
-     lw=3,
-     xlabel=L"\theta",
-     ylabel="Density")
-scatter!((mode(d2), pdf(d, mode(d2))), mc=:green, ms=5)
-plot!(range(q25, stop=q75, length=100),
-      x -> pdf(d, x),
-      lc=false, fc=:blues,
-      fill=true, fa=0.5)
-savefig(joinpath(@OUTPUT, "mixture.svg")); # hide
+credint = range(q25; stop=q75, length=100)
+
+f, ax, l = lines(
+    range_d,
+    pdf.(d, range_d);
+    linewidth=3,
+    axis=(;
+        limits=(-2, 14, nothing, nothing),
+        xticks=[0, 5, 10],
+        xlabel=L"\theta",
+        ylabel="Density",
+    ),
+)
+scatter!(ax, mode(d2), pdf(d, mode(d2)); color=:green, markersize=12)
+band!(ax, credint, 0.0, pdf.(d, credint); color=(:steelblue, 0.5))
+save(joinpath(@OUTPUT, "mixture.svg"), f); # hide
 
 # \fig{mixture}
 # \center{_**Mixture**: Maximum Likelihood Estimate vs Credible Intervals_} \\