Add a perceptron classifier to examples

ROADMAP.md

@@ -16,7 +16,7 @@
   - [x] regression
   - [ ] classification
   - [ ] clustering
-  - [ ] gradient descent
+  - [x] gradient descent
   - [x] iterative algorithms
   - [ ] incremental algorithms
   - [ ] dimension reduction

docs/src/common_implementation_patterns.md

@@ -22,12 +22,12 @@ implementations fall into one (or more) of the following informally understood p
 
 - [Regression](@ref): Supervised learners for continuous targets
 
-- Classification: Supervised learners for categorical targets 
+- [Classification](@ref): Supervised learners for categorical targets 
 
 - Clusterering: Algorithms that group data into clusters for classification and
   possibly dimension reduction. May be true learners (generalize to new data) or static.
 
-- Gradient Descent: Including neural networks.
+- [Gradient Descent](@ref): Including neural networks.
 
 - [Iterative Algorithms](@ref)
 

docs/src/patterns/classification.md

@@ -1 +1,5 @@
 # Classification
+
+See these examples from tests:
+
+- [perceptron classifier](https://github.com/JuliaAI/LearnAPI.jl/blob/dev/test/integration/gradient_descent.jl)

docs/src/patterns/gradient_descent.md

@@ -0,0 +1,5 @@
+# Gradient Descent
+
+See [this
+example](https://github.com/JuliaAI/LearnAPI.jl/blob/dev/test/integration/gradient_descent.jl)
+from tests.

docs/src/patterns/iterative_algorithms.md

@@ -1,5 +1,7 @@
 # Iterative Algorithms
 
-See [this
-example](https://github.com/JuliaAI/LearnAPI.jl/blob/dev/test/integration/ensembling.jl)
-from tests.
+See these examples from tests:
+
+- [bagged ensembling](https://github.com/JuliaAI/LearnAPI.jl/blob/dev/test/integration/ensembling.jl)
+
+- [perceptron classifier](https://github.com/JuliaAI/LearnAPI.jl/blob/dev/test/integration/gradient_descent.jl)

test/integration/ensembling.jl

@@ -194,13 +194,14 @@ Xtest = Tables.subset(X, test)
     ŷ7 = predict(model, Xtest)
 
     # compare with cold restart:
-    model = fit(LearnAPI.clone(algorithm; n=7), Xtrain, y[train]; verbosity=0);
-    @test ŷ7 ≈ predict(model, Xtest)
+    model_cold = fit(LearnAPI.clone(algorithm; n=7), Xtrain, y[train]; verbosity=0);
+    @test ŷ7 ≈ predict(model_cold, Xtest)
 
-    # test cold restart if another hyperparameter is changed:
+    # test that we get a cold restart if another hyperparameter is changed:
     model2 = update(model, Xtrain, y[train]; atom=Ridge(0.05))
-    algorithm2 = LearnAPI.clone(LearnAPI.algorithm(model); atom=Ridge(0.05))
-    @test predict(model, Xtest) ≈ predict(model2, Xtest)
+    algorithm2 = Ensemble(Ridge(0.05); n=7, rng)
+    model_cold = fit(algorithm2, Xtrain, y[train]; verbosity=0)
+    @test predict(model2, Xtest) ≈ predict(model_cold, Xtest)
 
 end