Add precompilation (#21)

Project.toml

@@ -1,13 +1,14 @@
 name = "Mice"
 uuid = "d4678d24-b338-4f96-a2c8-a66549d61c16"
 authors = ["Tom Metherell <[email protected]> and contributors"]
-version = "0.3.1"
+version = "0.3.2"
 
 [deps]
 AxisArrays = "39de3d68-74b9-583c-8d2d-e117c070f3a9"
 CategoricalArrays = "324d7699-5711-5eae-9e2f-1d82baa6b597"
 Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"
 LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
+PrecompileTools = "aea7be01-6a6a-4083-8856-8a6e6704d82a"
 Printf = "de0858da-6303-5e67-8744-51eddeeeb8d7"
 Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
 RecipesBase = "3cdcf5f2-1ef4-517c-9805-6587b60abb01"
@@ -18,16 +19,20 @@ StatsModels = "3eaba693-59b7-5ba5-a881-562e759f1c8d"
 Tables = "bd369af6-aec1-5ad0-b16a-f7cc5008161c"
 
 [weakdeps]
+DataFrames = "a93c6f00-e57d-5684-b7b6-d8193f3e46c0"
 RCall = "6f49c342-dc21-5d91-9882-a32aef131414"
 
 [extensions]
+MiceDataFramesExt = "DataFrames"
 MiceRCallExt = "RCall"
 
 [compat]
 AxisArrays = "0.4"
 CategoricalArrays = "0.10"
+DataFrames = "1.6"
 Distributions = "0.25"
-RCall = "0.13"
+PrecompileTools = "1.2"
+RCall = "0.13, 0.14"
 RecipesBase = "1.3.1"
 Statistics = "1"
 StatsAPI = "1.7"

ext/MiceDataFramesExt.jl

@@ -0,0 +1,43 @@
+module MiceDataFramesExt
+    using CategoricalArrays: CategoricalArray, CategoricalPool, CategoricalValue
+    using DataFrames: DataFrame
+    using Mice: bindImputations, complete, listComplete, makeMethods, mice
+    using PrecompileTools: @compile_workload
+    using Random: rand, randperm
+
+    @compile_workload begin
+        catPool = CategoricalPool(["a", "b", "c"])
+        df = DataFrame(
+            a = Vector{Union{Missing, Int}}(randperm(20)),
+            b = Vector{Union{Missing, Float64}}(randperm(20)),
+            c = Vector{Union{Missing, String}}(rand(["a", "b", "c"], 20)),
+            d = Vector{Union{Missing, Bool}}(rand(Bool, 20)),
+            e = CategoricalArray{Union{Missing, Int}}(rand([1, 2, 3], 20)),
+            f = CategoricalArray{Union{Missing, String}}(rand(["a", "b", "c"], 20)),
+            g = Vector{Union{Missing, CategoricalValue}}(rand([CategoricalValue(catPool, 1), CategoricalValue(catPool, 2), CategoricalValue(catPool, 3)], 20))
+        )
+
+        for col in axes(df, 2)
+            df[rand(1:20, 1), col] .= missing
+        end
+
+        imputedDataPmm = mice(df, m = 1, iter = 1, progressReports = false)
+
+        meanMethods = makeMethods(df)
+        meanMethods["b"] = "mean"
+        imputedDataMean = mice(df, m = 1, iter = 1, methods = meanMethods, progressReports = false)
+
+        normMethods = meanMethods
+        normMethods["b"] = "norm"
+        imputedDataNorm = mice(df, m = 1, iter = 1, methods = normMethods, progressReports = false)
+
+        sampleMethods = normMethods
+        sampleMethods[:] .= "sample"
+        imputedDataSample = mice(df, m = 1, iter = 1, methods = sampleMethods, progressReports = false)
+
+        bindImputations(imputedDataPmm, imputedDataPmm)
+
+        complete(imputedDataPmm, 1)
+        listComplete(imputedDataPmm)
+    end
+end

src/Mice.jl

@@ -1,9 +1,10 @@
 module Mice
     # Dependencies
     using AxisArrays: axes, AxisArray, AxisMatrix, AxisVector
-    using CategoricalArrays: CategoricalArray, CategoricalValue, levels
+    using CategoricalArrays: CategoricalArray, CategoricalPool, CategoricalValue, levels
     using Distributions: ccdf, Chisq, FDist, Normal, TDist
     using LinearAlgebra: cholesky, Diagonal, diagm, eigen, inv, qr, rank, svd
+    using PrecompileTools: @compile_workload
     using Printf: @printf
     using Random: rand, randn, randperm
     import RecipesBase: plot
@@ -459,4 +460,6 @@ module Mice
     end
 
     export bindImputations, complete, findMissings, listComplete, makeMethods, makePredictorMatrix, mice, Mids, Mipo, Mira, pool, plot, with
+
+    include("precompile.jl")
 end

src/precompile.jl

@@ -0,0 +1,35 @@
+@compile_workload begin
+    catPool = CategoricalPool(["a", "b", "c"])
+    ct = (
+        a = Vector{Union{Missing, Int}}(randperm(20)),
+        b = Vector{Union{Missing, Float64}}(randperm(20)),
+        c = Vector{Union{Missing, String}}(rand(["a", "b", "c"], 20)),
+        d = Vector{Union{Missing, Bool}}(rand(Bool, 20)),
+        e = CategoricalArray{Union{Missing, Int}}(rand([1, 2, 3], 20)),
+        f = CategoricalArray{Union{Missing, String}}(rand(["a", "b", "c"], 20)),
+        g = Vector{Union{Missing, CategoricalValue}}(rand([CategoricalValue(catPool, 1), CategoricalValue(catPool, 2), CategoricalValue(catPool, 3)], 20))
+    )
+
+    for col in ct
+        col[rand(1:20, 1)] .= missing
+    end
+
+    imputedDataPmm = mice(ct, m = 1, iter = 1, progressReports = false)
+
+    meanMethods = makeMethods(ct)
+    meanMethods["b"] = "mean"
+    imputedDataMean = mice(ct, m = 1, iter = 1, methods = meanMethods, progressReports = false)
+
+    normMethods = meanMethods
+    normMethods["b"] = "norm"
+    imputedDataNorm = mice(ct, m = 1, iter = 1, methods = normMethods, progressReports = false)
+
+    sampleMethods = normMethods
+    sampleMethods[:] .= "sample"
+    imputedDataSample = mice(ct, m = 1, iter = 1, methods = sampleMethods, progressReports = false)
+
+    bindImputations(imputedDataPmm, imputedDataPmm)
+
+    complete(imputedDataPmm, 1)
+    listComplete(imputedDataPmm)
+end