Merge pull request #230 from kescobo/ordinations

Add ordinations (MDS and PCA)

Project.toml

@@ -9,6 +9,7 @@ DataValues = "e7dc6d0d-1eca-5fa6-8ad6-5aecde8b7ea5"
 Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"
 Interpolations = "a98d9a8b-a2ab-59e6-89dd-64a1c18fca59"
 KernelDensity = "5ab0869b-81aa-558d-bb23-cbf5423bbe9b"
+MultivariateStats = "6f286f6a-111f-5878-ab1e-185364afe411"
 Observables = "510215fc-4207-5dde-b226-833fc4488ee2"
 Plots = "91a5bcdd-55d7-5caf-9e0b-520d859cae80"
 RecipesBase = "3cdcf5f2-1ef4-517c-9805-6587b60abb01"
@@ -18,8 +19,9 @@ Tables = "bd369af6-aec1-5ad0-b16a-f7cc5008161c"
 Widgets = "cc8bc4a8-27d6-5769-a93b-9d913e69aa62"
 
 [compat]
-julia = "≥ 1.0"
-Plots = "≥ 0.25"
-Widgets = "≥ 0.5"
+MultivariateStats = "≥ 0.7"
 Observables = "≥ 0.2.2"
+Plots = "≥ 0.25"
 RecipesBase = "≥ 0.6"
+Widgets = "≥ 0.5"
+julia = "≥ 1.0"

README.md

@@ -18,6 +18,7 @@ This package is a drop-in replacement for Plots.jl that contains many statistica
     - marginalhist
     - corrplot/cornerplot
     - andrewsplot
+    - MDS plot
 
 It is thus slightly less lightweight, but has more functionality. Main documentation is found in the Plots.jl documentation (https://juliaplots.github.io).
 
@@ -374,3 +375,23 @@ Some statistical analysis is computed at the single subject level (for example t
 For more information please refer to the [README](https://github.com/piever/GroupedErrors.jl/blob/master/README.md).
 
 A GUI based on QML and the GR Plots.jl backend to simplify the use of StatsPlots.jl and GroupedErrors.jl even further can be found [here](https://github.com/piever/PlugAndPlot.jl) (usable but still in alpha stage).
+
+## Ordinations
+
+MDS from [`MultivariateStats.jl`](https://github.com/JuliaStats/MultivariateStats.jl)
+can be plotted as scatter plots.
+
+```julia
+using MultivariateStats, RDatasets, StatsPlots
+
+iris = dataset("datasets", "iris")
+X = convert(Matrix, iris[:, 1:4])
+M = fit(MDS, X'; maxoutdim=2)
+
+plot(M, group=iris.Species)
+```
+
+![MDS plot](https://user-images.githubusercontent.com/3502975/64883550-a6186600-d62d-11e9-8f6b-c5094abf5573.png)
+
+PCA will be added once the API in MultivariateStats is changed.
+See https://github.com/JuliaStats/MultivariateStats.jl/issues/109 and https://github.com/JuliaStats/MultivariateStats.jl/issues/95.

src/StatsPlots.jl

@@ -15,6 +15,8 @@ import Widgets: @nodeps
 import DataStructures: OrderedDict
 import Clustering: Hclust, nnodes
 using Interpolations
+import MultivariateStats: MDS, eigvals, projection, principalvars,
+                            principalratio, transform
 
 import KernelDensity
 @recipe f(k::KernelDensity.UnivariateKDE) = k.x, k.density
@@ -37,5 +39,6 @@ include("marginalhist.jl")
 include("bar.jl")
 include("dendrogram.jl")
 include("andrews.jl")
+include("ordinations.jl")
 
 end # module

src/ordinations.jl

@@ -0,0 +1,26 @@
+@recipe function f(mds::MDS{<:Real}; mds_axes=(1,2))
+    length(mds_axes) in [2,3] || throw(ArgumentError("Can only accept 2 or 3 mds axes"))
+    xax = mds_axes[1]
+    yax = mds_axes[2]
+    ev = eigvals(mds)
+    var_explained = [v / sum(ev) for v in ev]
+    tfm = collect(transform(mds)')
+
+    xlabel --> "MDS$xax"
+    ylabel --> "MDS$yax"
+    seriestype := :scatter
+    aspect_ratio --> 1
+
+    if length(mds_axes) == 3
+        zax = mds_axes[3]
+        zlabel --> "MDS$zax"
+        tfm[:,xax], tfm[:,yax], tfm[:,zax]
+    else
+        tfm[:,xax], tfm[:,yax]
+    end
+end
+
+#= This needs to wait on a different PCA API in MultivariateStats.jl
+@recipe function f(pca::PCA{<:Real}; pca_axes=(1,2))
+end
+=#