Removed BetaML from test environment and improved API for wrap(decisi…

…on tree) for plotting

src/Trees/AbstractTrees_BetaML_interface.jl

@@ -38,8 +38,13 @@ In case of a `BetaML/DecisionTree` this is typically a list of feature names as
 
 `wdc = wrap(dc, (featurenames = feature_names, ))`
 """
+
 wrap(node::DecisionNode, info::NamedTuple = NamedTuple()) = InfoNode(node, info)
 wrap(leaf::Leaf,         info::NamedTuple = NamedTuple()) = InfoLeaf(leaf, info)
+wrap(mod::DecisionTreeEstimator, info::NamedTuple = NamedTuple()) = wrap(mod.par.tree, info)
+wrap(m::Union{DecisionNode,Leaf,DecisionTreeEstimator};feature_names=[]) = wrap(m,(featurenames=feature_names,))
+
+
 
 
 #### Implementation of the `AbstractTrees`-interface
@@ -61,4 +66,13 @@ function AbstractTrees.printnode(io::IO, leaf::InfoLeaf)
     for p in leaf.leaf.predictions
         println(io, p)
     end
+end
+
+function show(io::IO,node::Union{InfoNode,InfoLeaf})
+    #print(io, "Is col $(question.column) $condition $(question.value) ?")
+    print(io, "A wrapped Decision Tree")
+end
+
+function show(io::IO, ::MIME"text/plain", node::Union{InfoNode,InfoLeaf})
+    print(io, "A wrapped Decision Tree")
 end

src/Trees/DecisionTrees.jl

@@ -231,6 +231,34 @@ Dict{String, Any}("job_is_regression" => 1, "fitted_records" => 6, "max_reached_
                         --> False:  -13.8
 --> False:  3.3999999999999995
 ```
+
+- Visualisation...
+
+You can either text-print or plot a decision tree..
+
+```julia
+julia> println(mod)
+DecisionTreeEstimator - A Decision Tree regressor (fitted on 6 records)
+Dict{String, Any}("job_is_regression" => 1, "fitted_records" => 6, "max_reached_depth" => 4, "avg_depth" => 3.25, "xndims" => 2)
+*** Printing Decision Tree: ***
+
+1. Is col 2 >= 18.0 ?
+--> True :
+                1.2. Is col 2 >= 31.0 ?
+                --> True :  -27.2
+                --> False:
+                        1.2.3. Is col 2 >= 20.5 ?
+                        --> True :  -17.450000000000003
+                        --> False:  -13.8
+--> False:  3.3999999999999995
+
+using Plots, TreeRecipe
+feature_names = ["Something", "Som else"]
+wrapped_tree  = wrap(dtree, feature_names = feature_names) # feature_names is otional
+plot(wrapped_tree)    
+````
+![DT plot](assets/dtplot.png) 
+
 """
 mutable struct DecisionTreeEstimator <: BetaMLSupervisedModel
     hpar::DTHyperParametersSet
@@ -758,13 +786,6 @@ function computeDepths(node::AbstractNode)
     return (mean(leafDepths),maximum(leafDepths))
 end
 
-function show(io::IO,question::Question)
-    condition = "=="
-    if isa(question.value, Number)
-        condition = ">="
-    end
-    print(io, "Is col $(question.column) $condition $(question.value) ?")
-end
 
 """
   print(node)

test/Project.toml

@@ -1,6 +1,5 @@
 [deps]
 AbstractTrees = "1520ce14-60c1-5f80-bbc7-55ef81b5835c"
-BetaML = "024491cd-cc6b-443e-8034-08ea7eb7db2b"
 CategoricalArrays = "324d7699-5711-5eae-9e2f-1d82baa6b597"
 Combinatorics = "861a8166-3701-5b0c-9a16-15d98fcdc6aa"
 DelimitedFiles = "8bb1440f-4735-579b-a4ab-409b98df4dab"

test/Trees_tests_additional.jl

@@ -4,7 +4,7 @@ import MLJBase
 const Mlj = MLJBase
 using StableRNGs
 #rng = StableRNG(123)
-using BetaML.Trees
+using BetaML
 
 println("*** Additional testing for the Testing Decision trees/Random Forest algorithms...")
 
@@ -54,11 +54,10 @@ tree = Tree()
 
     model = DecisionTreeEstimator()
     yhat_train = Trees.fit!(model, xtrain, ytrain)
-    dtree = model.par.tree
 
     println("--> add information about feature names")
-    feature_names = ["Color", "Intensity"]
-    wrapped_tree = Trees.wrap(dtree, (featurenames = feature_names, ))
+    feature_names = ["Color", "Size"]
+    wrapped_tree = wrap(model, feature_names = feature_names)
 
     println("--> plot the tree using the `TreeRecipe`")
     plt = plot(wrapped_tree)        # this calls automatically the `TreeRecipe`