Merge 7c69f5f into 14f3b54

test/modelmatrix.jl

@@ -6,13 +6,12 @@
 
     sparsetype = SparseMatrixCSC{Float64,Int}
 
-    d = DataFrame()
-    d[:y] = [1:4;]
-    d[:x1] = [5:8;]
-    d[:x2] = [9:12;]
-    d[:x3] = [13:16;]
-    d[:x4] = [17:20;]
-    d[:x1p] = CategoricalArray(d[:x1])
+    d = DataFrame(y = 1:4,
+                  x1 = 5:8,
+                  x2 = 9:12,
+                  x3 = 13:16,
+                  x4 = 17:20)
+    d.x1p = categorical(d.x1)
 
     d_orig = deepcopy(d)
 
@@ -64,7 +63,7 @@
     @test mm.m == [ones(4) x1 x2 x1.*x2]
     @test mm.m == ModelMatrix{sparsetype}(mf).m
 
-    d[:x1] = CategoricalArray(x1)
+    d.x1 = categorical(x1)
     x1e = [[0, 1, 0, 0] [0, 0, 1, 0] [0, 0, 0, 1]]
     f = @formula(y ~ 1 + x1 * x2)
     mf = ModelFrame(f, d)
@@ -81,7 +80,7 @@
     @test response(mf) == y''
 
     d = deepcopy(d_orig)
-    d[:x1] = CategoricalArray{Union{Missing, Float64}}(d[:x1])
+    d.x1 = CategoricalArray{Union{Missing, Float64}}(d.x1)
 
     f = @formula(y ~ 1 + x2 + x3 + x3*x2)
     mm = ModelMatrix(ModelFrame(f, d))
@@ -144,7 +143,7 @@
     @test size(mm_sub) == (3,3)
 
     ## Missing data
-    d[:x1m] = [5, 6, missing, 7]
+    d.x1m = [5, 6, missing, 7]
     mf = ModelFrame(@formula(y ~ 1 + x1m), d)
     mm = ModelMatrix(mf)
     @test mm.m[:, 2] == d[completecases(d), :x1m]
@@ -162,7 +161,7 @@
                       z = repeat([:e, :f], inner = 4))
         categorical!(d)
         cs = Dict([Pair(name, EffectsCoding()) for name in names(d)])
-        d[:n] = 1.:8
+        d.n = 1.:8
 
 
         ## No intercept
@@ -182,7 +181,7 @@
         ## promotion blocked when we block default model=StatisticalModel
         mf = ModelFrame(@formula(n ~ 0 + x), d, model=Nothing, contrasts=cs)
         mm = ModelMatrix(mf)
-        @test all(mm.m .== ifelse.(d[:x] .== :a, -1, 1))
+        @test all(mm.m .== ifelse.(d.x .== :a, -1, 1))
         @test coefnames(mf) == ["x: b"]
 
 
@@ -315,7 +314,7 @@
         @test coefnames(mf) == ["(Intercept)", "1 | x"]
 
         mf = ModelFrame(@formula(y ~ 0 + (1 | x)), d)
-        @test all(ModelMatrix(mf).m .== float.(1 .| d[:x]))
+        @test all(ModelMatrix(mf).m .== float.(1 .| d.x))
         @test coefnames(mf) == ["1 | x"]
     end
 

test/statsmodel.jl

@@ -53,24 +53,24 @@ Base.show(io::IO, m::DummyModTwo) = println(io, m.msg)
 @testset "stat model types" begin
 
     ## Test fitting
-    d = DataFrame()
-    d[:y] = [1:4;]
-    d[:x1] = Vector{Union{Missing, Int}}(5:8)
-    d[:x2] = [9:12;]
-    d[:x3] = [13:16;]
-    d[:x4] = [17:20;]
-    d[:x1p] = CategoricalArray{Union{Missing, Int}}(d[:x1])
+    d = DataFrame(y = 1:4,
+                  x1 = allowmissing(5:8),
+                  x2 = 9:12,
+                  x3 = 13:16,
+                  x4 = 17:20)
+
+    d.x1p = categorical(d.x1)
 
     f = @formula(y ~ x1 * x2)
     m = fit(DummyMod, f, d)
-    @test response(m) == Array(d[:y])
+    @test response(m) == Array(d.y)
 
     ## coefnames delegated to model frame by default
     @test coefnames(m) == coefnames(ModelFrame(f, d)) == ["(Intercept)", "x1", "x2", "x1 & x2"]
 
     ## test prediction method
     ## vanilla
-    @test predict(m) == [ ones(size(d,1)) Array(d[:x1]) Array(d[:x2]) Array(d[:x1]).*Array(d[:x2]) ] * collect(1:4)
+    @test predict(m) == [ ones(size(d,1)) Array(d.x1) Array(d.x2) Array(d.x1).*Array(d.x2) ] * collect(1:4)
 
     ## new data from matrix
     mm = ModelMatrix(ModelFrame(f, d))
@@ -103,7 +103,7 @@ Base.show(io::IO, m::DummyModTwo) = println(io, m.msg)
     ## predict w/ new data with _extra_ levels (throws an error)
     d3 = deepcopy(d)
     d3[1, :x1] = 0
-    d3[:x1p] = CategoricalVector{Union{Missing, Int}}(d3[:x1])
+    d3.x1p = categorical(d3.x1)
     # TODO: check for level mismatch earlier...this throws a KeyError when it
     # goes to do the lookup in the contrasts matrix from the previously
     # generated categorical term.
@@ -112,14 +112,14 @@ Base.show(io::IO, m::DummyModTwo) = println(io, m.msg)
 
     ## predict with dataframe that doesn't have the dependent variable
     d4 = deepcopy(d)
-    deletecols!(d4, [:y])
+    select!(d4, Not(:y))
     @test predict(m, d4) == predict(m, d)
 
     ## attempting to fit with d4 should fail since it doesn't have :y
     @test_throws ErrorException fit(DummyMod, f, d4)
 
     ## fit with contrasts specified
-    d[:x2p] = CategoricalVector{Union{Missing, Int}}(d[:x2])
+    d.x2p = categorical(d.x2)
     f3 = @formula(y ~ x1p + x2p)
     m3 = fit(DummyMod, f3, d)
     fit(DummyMod, f3, d, contrasts = Dict(:x1p => EffectsCoding()))