.github/workflows/CI.yml

@@ -18,7 +18,7 @@ jobs:
       fail-fast: false
       matrix:
         version:
-          - '1.9'
+          - 'lts'
           - '1'
         os:
           - ubuntu-latest
@@ -45,7 +45,7 @@ jobs:
       - uses: julia-actions/julia-buildpkg@v1
       - uses: julia-actions/julia-runtest@v1
       - uses: julia-actions/julia-processcoverage@v1
-      - uses: codecov/codecov-action@v4
+      - uses: codecov/codecov-action@v5
         with:
           file: lcov.info
           token: ${{ secrets.CODECOV_TOKEN }}

.github/workflows/FormatPR.yml

@@ -10,11 +10,11 @@ jobs:
       - uses: actions/checkout@v4
       - name: Install JuliaFormatter and format
         run: |
-          julia -e 'import Pkg; Pkg.add("JuliaFormatter")'
+          julia -e 'import Pkg; Pkg.add(name="JuliaFormatter", version="1")'
           julia -e 'using JuliaFormatter; format(".")'
       - name: Create Pull Request
         id: cpr
-        uses: peter-evans/create-pull-request@v6
+        uses: peter-evans/create-pull-request@v7
         with:
           token: ${{ secrets.GITHUB_TOKEN }}
           commit-message: ":robot: Format .jl files"

Project.toml

@@ -1,7 +1,7 @@
 name = "TableTransforms"
 uuid = "0d432bfd-3ee1-4ac1-886a-39f05cc69a3e"
 authors = ["Júlio Hoffimann <julio.hoffimann@gmail.com> and contributors"]
-version = "1.33.0"
+version = "1.34.1"
 
 [deps]
 AbstractTrees = "1520ce14-60c1-5f80-bbc7-55ef81b5835c"
@@ -18,6 +18,7 @@ PrettyTables = "08abe8d2-0d0c-5749-adfa-8a2ac140af0d"
 Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
 Statistics = "10745b16-79ce-11e8-11f9-7d13ad32a3b2"
 StatsBase = "2913bbd2-ae8a-5f71-8c99-4fb6c76f3a91"
+TableDistances = "e5d66e97-8c70-46bb-8b66-04a2d73ad782"
 Tables = "bd369af6-aec1-5ad0-b16a-f7cc5008161c"
 TransformsBase = "28dd2a49-a57a-4bfb-84ca-1a49db9b96b8"
 Unitful = "1986cc42-f94f-5a68-af5c-568840ba703d"
@@ -26,8 +27,8 @@ Unitful = "1986cc42-f94f-5a68-af5c-568840ba703d"
 AbstractTrees = "0.4"
 CategoricalArrays = "0.10"
 CoDa = "1.2"
-ColumnSelectors = "0.1"
-DataScienceTraits = "0.3"
+ColumnSelectors = "1.0"
+DataScienceTraits = "1.0"
 Distributed = "1.9"
 Distributions = "0.25"
 InverseFunctions = "0.1"
@@ -37,6 +38,7 @@ PrettyTables = "2"
 Random = "1.9"
 Statistics = "1.9"
 StatsBase = "0.33, 0.34"
+TableDistances = "1.0"
 Tables = "1.6"
 TransformsBase = "1.5"
 Unitful = "1.17"

docs/src/transforms.md

@@ -242,6 +242,12 @@ SDS
 ProjectionPursuit
 ```
 
+## KMedoids
+
+```@docs
+KMedoids
+```
+
 ## Closure
 
 ```@docs

src/TableTransforms.jl

@@ -6,19 +6,20 @@ module TableTransforms
 
 using Tables
 using Unitful
-using Statistics
 using PrettyTables
 using AbstractTrees
-using LinearAlgebra
+using TableDistances
 using DataScienceTraits
 using CategoricalArrays
+using LinearAlgebra
+using Statistics
 using Random
 using CoDa
 
+using DataScienceTraits: SciType, coerce
 using TransformsBase: Transform, Identity, →
 using ColumnSelectors: ColumnSelector, SingleColumnSelector
 using ColumnSelectors: AllSelector, Column, selector, selectsingle
-using DataScienceTraits: SciType, Continuous, Categorical, coerce
 using Unitful: AbstractQuantity, AffineQuantity, AffineUnits, Units
 using Distributions: ContinuousUnivariateDistribution, Normal
 using InverseFunctions: NoInverse, inverse as invfun
@@ -90,6 +91,7 @@ export
   DRS,
   SDS,
   ProjectionPursuit,
+  KMedoids,
   Closure,
   Remainder,
   Compose,

src/distributions.jl

@@ -20,7 +20,7 @@ EmpiricalDistribution(values) = EmpiricalDistribution{eltype(values)}(values)
 
 quantile(d::EmpiricalDistribution, p::Real) = quantile(d.values, p, sorted=true)
 
-function cdf(d::EmpiricalDistribution{T}, x::T) where {T}
+function cdf(d::EmpiricalDistribution, x)
   v = d.values
   n = length(v)
 
@@ -37,9 +37,9 @@ function cdf(d::EmpiricalDistribution{T}, x::T) where {T}
   l, u = v[head], v[tail]
 
   if x < l
-    return T(0)
+    return 0.0
   elseif x > u
-    return T(1)
+    return 1.0
   else
     if l == u
       return tail / n

src/transforms.jl

@@ -286,6 +286,7 @@ include("transforms/quantile.jl")
 include("transforms/functional.jl")
 include("transforms/eigenanalysis.jl")
 include("transforms/projectionpursuit.jl")
+include("transforms/kmedoids.jl")
 include("transforms/closure.jl")
 include("transforms/remainder.jl")
 include("transforms/compose.jl")

src/transforms/closure.jl

@@ -13,19 +13,17 @@ See also [`Remainder`](@ref).
 """
 struct Closure <: StatelessFeatureTransform end
 
-isrevertible(::Type{Closure}) = true
-
 assertions(::Closure) = [scitypeassert(Continuous)]
 
 function applyfeat(::Closure, feat, prep)
   cols = Tables.columns(feat)
   names = Tables.columnnames(cols)
 
-  # table as matrix and get the sum acros dims 2
+  # convert table to matrix
   X = Tables.matrix(feat)
-  S = sum(X, dims=2)
 
-  # divides each row by its sum (closure operation)
+  # divide each row by its sum (closure operation)
+  S = sum(X, dims=2)
   Z = X ./ S
 
   # table with the old columns and the new values
@@ -34,21 +32,3 @@ function applyfeat(::Closure, feat, prep)
 
   newfeat, S
 end
-
-function revertfeat(::Closure, newfeat, fcache)
-  cols = Tables.columns(newfeat)
-  names = Tables.columnnames(cols)
-
-  # table as matrix
-  Z = Tables.matrix(newfeat)
-
-  # retrieve cache
-  S = fcache
-
-  # undo operation
-  X = Z .* S
-
-  # table with original columns
-  𝒯 = (; zip(names, eachcol(X))...)
-  𝒯 |> Tables.materializer(newfeat)
-end

src/transforms/coalesce.jl

@@ -44,8 +44,6 @@ Coalesce(cols::C...; value) where {C<:Column} = Coalesce(selector(cols), value)
 
 parameters(transform::Coalesce) = (; value=transform.value)
 
-isrevertible(::Type{<:Coalesce}) = false
-
 colcache(::Coalesce, x) = nothing
 
 colapply(transform::Coalesce, x, c) = coalesce.(x, transform.value)

src/transforms/coerce.jl

@@ -16,10 +16,10 @@ This transform uses the `DataScienceTraits.coerce` function. Please see their do
 # Examples
 
 ```julia
-import DataScienceTraits as DST
-Coerce(1 => DST.Continuous, 2 => DST.Continuous)
-Coerce(:a => DST.Continuous, :b => DST.Continuous)
-Coerce("a" => DST.Continuous, "b" => DST.Continuous)
+using DataScienceTraits
+Coerce(1 => Continuous, 2 => Continuous)
+Coerce(:a => Continuous, :b => Continuous)
+Coerce("a" => Continuous, "b" => Continuous)
 ```
 """
 struct Coerce{S<:ColumnSelector,T} <: StatelessFeatureTransform

src/transforms/compose.jl

@@ -3,37 +3,37 @@
 # ------------------------------------------------------------------
 
 """
-    Compose(; as=:CODA)
+    Compose(; as=:coda)
 
 Converts all columns of the table into parts of a composition
 in a new column named `as`, using the `CoDa.compose` function.
 
-    Compose(col₁, col₂, ..., colₙ; as=:CODA)
-    Compose([col₁, col₂, ..., colₙ]; as=:CODA)
-    Compose((col₁, col₂, ..., colₙ); as=:CODA)
+    Compose(col₁, col₂, ..., colₙ; as=:coda)
+    Compose([col₁, col₂, ..., colₙ]; as=:coda)
+    Compose((col₁, col₂, ..., colₙ); as=:coda)
 
 Converts the selected columns `col₁`, `col₂`, ..., `colₙ` into parts of a composition.
 
-    Compose(regex; as=:CODA)
+    Compose(regex; as=:coda)
 
 Converts the columns that match with `regex` into parts of a composition.
 
 # Examples
 
 ```julia
-Compose(as=:comp)
+Compose(as=:composition)
 Compose([2, 3, 5])
 Compose([:b, :c, :e])
 Compose(("b", "c", "e"))
-Compose(r"[bce]", as="COMP")
+Compose(r"[bce]", as="composition")
 ```
 """
 struct Compose{S<:ColumnSelector} <: StatelessFeatureTransform
   selector::S
   as::Symbol
 end
 
-Compose(selector::ColumnSelector; as=:CODA) = Compose(selector, Symbol(as))
+Compose(selector::ColumnSelector; as=:coda) = Compose(selector, Symbol(as))
 
 Compose(; kwargs...) = Compose(AllSelector(); kwargs...)
 Compose(cols; kwargs...) = Compose(selector(cols); kwargs...)

src/transforms/dropextrema.jl

@@ -52,8 +52,6 @@ DropExtrema(cols::C...; low=0.25, high=0.75) where {C<:Column} = DropExtrema(sel
 
 parameters(transform::DropExtrema) = (low=transform.low, high=transform.high)
 
-isrevertible(::Type{<:DropExtrema}) = false
-
 function preprocess(transform::DropExtrema, feat)
   cols = Tables.columns(feat)
   names = Tables.columnnames(cols)

src/transforms/dropmissing.jl

@@ -41,8 +41,6 @@ DropMissing() = DropMissing(AllSelector())
 DropMissing(cols) = DropMissing(selector(cols))
 DropMissing(cols::C...) where {C<:Column} = DropMissing(selector(cols))
 
-isrevertible(::Type{<:DropMissing}) = false
-
 function preprocess(transform::DropMissing, feat)
   cols = Tables.columns(feat)
   names = Tables.columnnames(cols)

src/transforms/dropnan.jl

@@ -35,8 +35,6 @@ DropNaN() = DropNaN(AllSelector())
 DropNaN(cols) = DropNaN(selector(cols))
 DropNaN(cols::C...) where {C<:Column} = DropNaN(selector(cols))
 
-isrevertible(::Type{<:DropNaN}) = false
-
 _isnan(_) = false
 _isnan(x::Number) = isnan(x)
 

src/transforms/filter.jl

@@ -27,8 +27,6 @@ struct Filter{F} <: StatelessFeatureTransform
   pred::F
 end
 
-isrevertible(::Type{<:Filter}) = false
-
 function preprocess(transform::Filter, feat)
   # lazy row iterator
   rows = tablerows(feat)