From a4c8610b967d9938ea8d87e4991260428e7d5315 Mon Sep 17 00:00:00 2001
From: Mateusz Baran <mateuszbaran89@gmail.com>
Date: Mon, 19 Oct 2020 16:12:50 +0200
Subject: [PATCH 1/7] support multivariate kNN regression

---
 src/NearestNeighbors.jl  |  2 +-
 test/NearestNeighbors.jl | 14 ++++++++++++++
 2 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/src/NearestNeighbors.jl b/src/NearestNeighbors.jl
index 023a89a9..110c9ec6 100644
--- a/src/NearestNeighbors.jl
+++ b/src/NearestNeighbors.jl
@@ -128,7 +128,7 @@ end
 function MMI.predict(m::KNNRegressor, (tree, y, w), X)
     Xmatrix     = MMI.matrix(X, transpose=true) # NOTE: copies the data
     idxs, dists = NN.knn(tree, Xmatrix, m.K)
-    preds       = zeros(length(idxs))
+    preds       = similar(y, length(idxs))
 
     w_ = ones(m.K)
 
diff --git a/test/NearestNeighbors.jl b/test/NearestNeighbors.jl
index 30dca37f..020f7ab2 100644
--- a/test/NearestNeighbors.jl
+++ b/test/NearestNeighbors.jl
@@ -109,6 +109,20 @@ p2 = predict(knnr, f2, xtest)
 @test all(p[ntest+1:2*ntest] .≈ 2.0)
 @test all(p[2*ntest+1:end] .≈ -2.0)
 
+y1v = fill( [0.0], n)
+y2v = fill( [2.0], n)
+y3v = fill([-2.0], n)
+
+yv = vcat(y1v, y2v, y3v)
+
+fv,_,_ = fit(knnr, 1, x, yv)
+f2v,_,_ = fit(knnr, 1, x, yv, w)
+
+pv = predict(knnr, fv, xtest)
+
+@test all(pv[1:ntest] .≈ [[0.0]])
+@test all(pv[ntest+1:2*ntest] .≈ [[2.0]])
+@test all(pv[2*ntest+1:end] .≈ [[-2.0]])
 
 
 

From 0b8c3bd4d613862acd57b40df2c16ada7bde8f0e Mon Sep 17 00:00:00 2001
From: Mateusz Baran <mateuszbaran89@gmail.com>
Date: Tue, 20 Oct 2020 20:37:36 +0200
Subject: [PATCH 2/7] updated target of kNN regressor

---
 src/NearestNeighbors.jl  | 2 +-
 test/NearestNeighbors.jl | 3 +--
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/src/NearestNeighbors.jl b/src/NearestNeighbors.jl
index 110c9ec6..65f7b848 100644
--- a/src/NearestNeighbors.jl
+++ b/src/NearestNeighbors.jl
@@ -161,7 +161,7 @@ metadata_pkg.((KNNRegressor, KNNClassifier),
 
 metadata_model(KNNRegressor,
     input   = Table(Continuous),
-    target  = AbstractVector{Continuous},
+    target  = Union{Table(Continuous), AbstractVector{Continuous}},
     weights = true,
     descr   = KNNRegressorDescription
     )
diff --git a/test/NearestNeighbors.jl b/test/NearestNeighbors.jl
index 020f7ab2..e62a5b35 100644
--- a/test/NearestNeighbors.jl
+++ b/test/NearestNeighbors.jl
@@ -142,8 +142,7 @@ infos[:docstring]
 infos = info_dict(knnr)
 
 @test infos[:input_scitype] == Table(Continuous)
-@test infos[:target_scitype] == AbstractVector{Continuous}
-
+@test infos[:target_scitype] == Union{Table(Continuous), AbstractVector{Continuous}}
 infos[:docstring]
 
 end

From 865a35200e858aa396f0dc807b7bcc568f0dd6e2 Mon Sep 17 00:00:00 2001
From: Mateusz Baran <mateuszbaran89@gmail.com>
Date: Tue, 20 Oct 2020 21:39:33 +0200
Subject: [PATCH 3/7] changing target of KNNRegressor

---
 src/NearestNeighbors.jl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/NearestNeighbors.jl b/src/NearestNeighbors.jl
index 65f7b848..9ee57abc 100644
--- a/src/NearestNeighbors.jl
+++ b/src/NearestNeighbors.jl
@@ -161,7 +161,7 @@ metadata_pkg.((KNNRegressor, KNNClassifier),
 
 metadata_model(KNNRegressor,
     input   = Table(Continuous),
-    target  = Union{Table(Continuous), AbstractVector{Continuous}},
+    target  = Union{AbstractVector{Continuous}, AbstractVector{<:AbstractVector{Continuous}}},
     weights = true,
     descr   = KNNRegressorDescription
     )

From 3062a05d3de93298337b34b4abe371d7102b0d34 Mon Sep 17 00:00:00 2001
From: Mateusz Baran <mateuszbaran89@gmail.com>
Date: Tue, 20 Oct 2020 21:43:43 +0200
Subject: [PATCH 4/7] target of kNN regressor again

---
 src/NearestNeighbors.jl  | 2 +-
 test/NearestNeighbors.jl | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/NearestNeighbors.jl b/src/NearestNeighbors.jl
index 9ee57abc..ae6809ff 100644
--- a/src/NearestNeighbors.jl
+++ b/src/NearestNeighbors.jl
@@ -161,7 +161,7 @@ metadata_pkg.((KNNRegressor, KNNClassifier),
 
 metadata_model(KNNRegressor,
     input   = Table(Continuous),
-    target  = Union{AbstractVector{Continuous}, AbstractVector{<:AbstractVector{Continuous}}},
+    target  = Union{AbstractVector{Continuous}, AbstractVector{<:AbstractArray{Continuous}}},
     weights = true,
     descr   = KNNRegressorDescription
     )
diff --git a/test/NearestNeighbors.jl b/test/NearestNeighbors.jl
index e62a5b35..c5cfdb9c 100644
--- a/test/NearestNeighbors.jl
+++ b/test/NearestNeighbors.jl
@@ -142,7 +142,7 @@ infos[:docstring]
 infos = info_dict(knnr)
 
 @test infos[:input_scitype] == Table(Continuous)
-@test infos[:target_scitype] == Union{Table(Continuous), AbstractVector{Continuous}}
+@test infos[:target_scitype] == Union{AbstractVector{Continuous}, AbstractVector{<:AbstractArray{Continuous}}}
 infos[:docstring]
 
 end

From e52ea0aafaef89543159b608528c9f9d120926b1 Mon Sep 17 00:00:00 2001
From: Mateusz Baran <mateuszbaran89@gmail.com>
Date: Wed, 21 Oct 2020 15:44:30 +0200
Subject: [PATCH 5/7] trying to make the multi-target kNN regressor work with
 tables

---
 src/NearestNeighbors.jl  | 26 ++++++++++++++++++++------
 test/NearestNeighbors.jl | 16 +++++++---------
 2 files changed, 27 insertions(+), 15 deletions(-)

diff --git a/src/NearestNeighbors.jl b/src/NearestNeighbors.jl
index ae6809ff..3b810e1d 100644
--- a/src/NearestNeighbors.jl
+++ b/src/NearestNeighbors.jl
@@ -128,26 +128,40 @@ end
 function MMI.predict(m::KNNRegressor, (tree, y, w), X)
     Xmatrix     = MMI.matrix(X, transpose=true) # NOTE: copies the data
     idxs, dists = NN.knn(tree, Xmatrix, m.K)
-    preds       = similar(y, length(idxs))
+
+    if typeof(y) <: AbstractVector
+        ymat    = reshape(y, length(y), 1)
+        preds   = similar(ymat, length(idxs), 1)
+    else # for multi-target prediction
+        ymat    = MMI.matrix(y)
+        preds   = similar(ymat, length(idxs), size(ymat, 2))
+    end
 
     w_ = ones(m.K)
 
     for i in eachindex(idxs)
         idxs_  = idxs[i]
+        println(idxs_)
         dists_ = dists[i]
-        values = y[idxs_]
+        values = [ymat[j,:] for j in idxs_]
         if w !== nothing
             w_ = w[idxs_]
         end
+        println(preds)
         if m.weights == :uniform
-            preds[i] = sum(values .* w_) / sum(w_)
+            preds[i,:] .= sum(values .* w_) / sum(w_)
         else
-            preds[i] = sum(values .* w_ .* (1.0 .- dists_ ./ sum(dists_))) / (sum(w_) - 1)
+            preds[i,:] .= sum(values .* w_ .* (1.0 .- dists_ ./ sum(dists_))) / (sum(w_) - 1)
         end
     end
-    return preds
+    if typeof(x) <: AbstractArray
+        return preds
+    else
+        return MMI.table(preds, names=Tables.schema(y).names, prototype=y)
+    end
 end
 
+
 # ====
 
 metadata_pkg.((KNNRegressor, KNNClassifier),
@@ -161,7 +175,7 @@ metadata_pkg.((KNNRegressor, KNNClassifier),
 
 metadata_model(KNNRegressor,
     input   = Table(Continuous),
-    target  = Union{AbstractVector{Continuous}, AbstractVector{<:AbstractArray{Continuous}}},
+    target  = Union{AbstractVector{Continuous}, Table{Continuous}},
     weights = true,
     descr   = KNNRegressorDescription
     )
diff --git a/test/NearestNeighbors.jl b/test/NearestNeighbors.jl
index c5cfdb9c..5d39dbec 100644
--- a/test/NearestNeighbors.jl
+++ b/test/NearestNeighbors.jl
@@ -109,21 +109,19 @@ p2 = predict(knnr, f2, xtest)
 @test all(p[ntest+1:2*ntest] .≈ 2.0)
 @test all(p[2*ntest+1:end] .≈ -2.0)
 
-y1v = fill( [0.0], n)
-y2v = fill( [2.0], n)
-y3v = fill([-2.0], n)
-
-yv = vcat(y1v, y2v, y3v)
+ymat = vcat(fill( 0.0, n, 2), fill(2.0, n, 2), fill(-2.0, n, 2))
+yv = Tables.table(ymat; header = [:a, :b])
 
 fv,_,_ = fit(knnr, 1, x, yv)
 f2v,_,_ = fit(knnr, 1, x, yv, w)
 
 pv = predict(knnr, fv, xtest)
 
-@test all(pv[1:ntest] .≈ [[0.0]])
-@test all(pv[ntest+1:2*ntest] .≈ [[2.0]])
-@test all(pv[2*ntest+1:end] .≈ [[-2.0]])
-
+for col in [:a, :b]
+    @test all(pv[col][1:ntest] .≈ [0.0])
+    @test all(pv[col][ntest+1:2*ntest] .≈ [2.0])
+    @test all(pv[col][2*ntest+1:end] .≈ [-2.0])
+end
 
 
 

From a794ade31464a255a10a228aa102f529a209c23a Mon Sep 17 00:00:00 2001
From: Mateusz Baran <mateuszbaran89@gmail.com>
Date: Wed, 21 Oct 2020 20:04:01 +0200
Subject: [PATCH 6/7] fixing kNN regressor

---
 src/NearestNeighbors.jl  | 5 ++---
 test/NearestNeighbors.jl | 3 ++-
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/src/NearestNeighbors.jl b/src/NearestNeighbors.jl
index 3b810e1d..ea15ebb4 100644
--- a/src/NearestNeighbors.jl
+++ b/src/NearestNeighbors.jl
@@ -10,6 +10,7 @@ const MMI = MLJModelInterface
 using Distances
 
 import ..NearestNeighbors
+import ..Tables
 
 const NN = NearestNeighbors
 
@@ -141,20 +142,18 @@ function MMI.predict(m::KNNRegressor, (tree, y, w), X)
 
     for i in eachindex(idxs)
         idxs_  = idxs[i]
-        println(idxs_)
         dists_ = dists[i]
         values = [ymat[j,:] for j in idxs_]
         if w !== nothing
             w_ = w[idxs_]
         end
-        println(preds)
         if m.weights == :uniform
             preds[i,:] .= sum(values .* w_) / sum(w_)
         else
             preds[i,:] .= sum(values .* w_ .* (1.0 .- dists_ ./ sum(dists_))) / (sum(w_) - 1)
         end
     end
-    if typeof(x) <: AbstractArray
+    if typeof(y) <: AbstractArray
         return preds
     else
         return MMI.table(preds, names=Tables.schema(y).names, prototype=y)
diff --git a/test/NearestNeighbors.jl b/test/NearestNeighbors.jl
index 5d39dbec..6ae0a8a0 100644
--- a/test/NearestNeighbors.jl
+++ b/test/NearestNeighbors.jl
@@ -7,6 +7,7 @@ using MLJModels.NearestNeighbors_
 using CategoricalArrays
 using MLJBase
 using Random
+using Tables
 
 Random.seed!(5151)
 
@@ -140,7 +141,7 @@ infos[:docstring]
 infos = info_dict(knnr)
 
 @test infos[:input_scitype] == Table(Continuous)
-@test infos[:target_scitype] == Union{AbstractVector{Continuous}, AbstractVector{<:AbstractArray{Continuous}}}
+@test infos[:target_scitype] == Union{AbstractVector{Continuous}, Table{Continuous}}
 infos[:docstring]
 
 end

From 6f280403fe0816fc30cacb760cee5c27ca1b2ab2 Mon Sep 17 00:00:00 2001
From: Mateusz Baran <mateuszbaran89@gmail.com>
Date: Thu, 22 Oct 2020 10:20:37 +0200
Subject: [PATCH 7/7] code review fixes

---
 src/NearestNeighbors.jl  | 4 ++--
 test/NearestNeighbors.jl | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/src/NearestNeighbors.jl b/src/NearestNeighbors.jl
index ea15ebb4..60b31753 100644
--- a/src/NearestNeighbors.jl
+++ b/src/NearestNeighbors.jl
@@ -153,7 +153,7 @@ function MMI.predict(m::KNNRegressor, (tree, y, w), X)
             preds[i,:] .= sum(values .* w_ .* (1.0 .- dists_ ./ sum(dists_))) / (sum(w_) - 1)
         end
     end
-    if typeof(y) <: AbstractArray
+    if typeof(y) <: AbstractVector
         return preds
     else
         return MMI.table(preds, names=Tables.schema(y).names, prototype=y)
@@ -174,7 +174,7 @@ metadata_pkg.((KNNRegressor, KNNClassifier),
 
 metadata_model(KNNRegressor,
     input   = Table(Continuous),
-    target  = Union{AbstractVector{Continuous}, Table{Continuous}},
+    target  = Union{AbstractVector{Continuous}, Table(Continuous)},
     weights = true,
     descr   = KNNRegressorDescription
     )
diff --git a/test/NearestNeighbors.jl b/test/NearestNeighbors.jl
index 6ae0a8a0..4048499a 100644
--- a/test/NearestNeighbors.jl
+++ b/test/NearestNeighbors.jl
@@ -141,7 +141,7 @@ infos[:docstring]
 infos = info_dict(knnr)
 
 @test infos[:input_scitype] == Table(Continuous)
-@test infos[:target_scitype] == Union{AbstractVector{Continuous}, Table{Continuous}}
+@test infos[:target_scitype] == Union{AbstractVector{Continuous}, Table(Continuous)}
 infos[:docstring]
 
 end