Fix show method for PCA

.github/workflows/ci.yml

@@ -14,22 +14,17 @@ jobs:
       fail-fast: false
       matrix:
         version:
-          - '1.1'
-          - '1.6' # current LTS release
-          - '1' # automatically expands to the latest stable 1.x release of Julia
-          # - 'nightly'
+          - 'min'
+          - 'lts'
+          - '1'
+          - 'pre'
         os:
           - ubuntu-latest
           # - macOS-latest
           # - windows-latest
         arch:
           - x64
         allow-to-fail: [false]
-        include:
-          - version: 'nightly'
-            os: ubuntu-latest
-            arch: x64
-            allow-to-fail: true
     steps:
       - uses: actions/checkout@v6
       - uses: julia-actions/setup-julia@v2

Project.toml

@@ -4,7 +4,7 @@ keywords = ["multivariate statistics", "dimensionality reduction"]
 license = "MIT"
 desc = "A Julia package for multivariate statistics and data analysis"
 repository = "https://github.com/JuliaStats/MultivariateStats.jl.git"
-version = "0.10.3"
+version = "0.11.0"
 
 [deps]
 Arpack = "7d9fca2a-8960-54d3-9f78-7d1dccf2cb97"
@@ -22,7 +22,7 @@ StableRNGs = "1"
 Statistics = "1"
 StatsAPI = "^1.3"
 StatsBase = "^0.33, 0.34"
-julia = "1.1"
+julia = "1.6"
 
 [extras]
 StableRNGs = "860ef19b-820b-49d6-a774-d7a799459cd3"

src/pca.jl

@@ -153,7 +153,7 @@ function show(io::IO, ::MIME"text/plain", M::PCA)
     ldgs = ldgs * diagm(0 => ldgs_signs[:])
     print(io, "\n\nPattern matrix (unstandardized loadings):\n")
     cft = CoefTable(ldgs, string.("PC", 1:odim), string.("", 1:idim))
-    print(io, cft)
+    show(io, MIME"text/plain"(), cft)
     print(io, "\n\n")
     print(io, "Importance of components:\n")
     λ = eigvals(M)
@@ -164,7 +164,7 @@ function show(io::IO, ::MIME"text/plain", M::PCA)
              "Proportion explained", "Cumulative proportion"]
     cft = CoefTable(vcat(λ', prp', cumsum(prp)',  prpv', cumsum(prpv)'),
                     string.("PC", 1:odim), names)
-    print(io, cft)
+    show(io, MIME"text/plain"(), cft)
 end
 
 #### PCA Training

test/pca.jl

@@ -43,6 +43,30 @@ import SparseArrays
     @test reconstruct(M, Y[:,1]) ≈ P * Y[:,1]
     @test reconstruct(M, Y) ≈ P * Y
 
+    @test repr("text/plain", M) == """
+        PCA(indim = 5, outdim = 3, principalratio = 0.8)
+
+        Pattern matrix (unstandardized loadings):
+        ─────────────────────────────────
+                 PC1        PC2       PC3
+        ─────────────────────────────────
+        1  -0.236399   0.706558  1.21106
+        2   1.27426    0.047799  0.212282
+        3   0.990598  -1.22718   0.882179
+        4  -1.51861   -0.725771  0.631665
+        5   0.181386   1.21069   0.557706
+        ─────────────────────────────────
+
+        Importance of components:
+        ───────────────────────────────────────────────────
+                                        PC1       PC2   PC3
+        ───────────────────────────────────────────────────
+        SS Loadings (Eigenvalues)  5.0       4.0       3.0
+        Variance explained         0.333333  0.266667  0.2
+        Cumulative variance        0.333333  0.6       0.8
+        Proportion explained       0.416667  0.333333  0.25
+        Cumulative proportion      0.416667  0.75      1.0
+        ───────────────────────────────────────────────────"""
 
     ## PCA with non-zero mean