Merge pull request #183 from JuliaAI/dont-export-measures

Stop exporting measures

Author: ablaom

README.md

@@ -92,7 +92,7 @@ apply_tree(model, [5.9,3.0,5.1,1.9])
 # apply model to all the sames
 preds = apply_tree(model, features)
 # generate confusion matrix, along with accuracy and kappa scores
-confusion_matrix(labels, preds)
+DecisionTree.confusion_matrix(labels, preds)
 # get the probability of each label
 apply_tree_proba(model, [5.9,3.0,5.1,1.9], ["Iris-setosa", "Iris-versicolor", "Iris-virginica"])
 # run 3-fold cross validation of pruned tree,
@@ -312,6 +312,13 @@ Available models are: `AdaBoostStumpClassifier`,
 `RandomForestClassifier`, `RandomForestRegressor`.
 
 
+## Feature Importances
+
+The following methods provide measures of feature importance for all models:
+`impurity_importance`, `split_importance`, `permutation_importance`. Query the document
+strings for details.
+
+
 ## Saving Models
 Models can be saved to disk and loaded back with the use of the [JLD2.jl](https://github.com/JuliaIO/JLD2.jl) package.
 ```julia

src/DecisionTree.jl

@@ -1,5 +1,3 @@
-__precompile__()
-
 module DecisionTree
 
 import Base: length, show, convert, promote_rule, zero
@@ -13,8 +11,7 @@ export Leaf, Node, Root, Ensemble, print_tree, depth, build_stump, build_tree,
        prune_tree, apply_tree, apply_tree_proba, nfoldCV_tree, build_forest,
        apply_forest, apply_forest_proba, nfoldCV_forest, build_adaboost_stumps,
        apply_adaboost_stumps, apply_adaboost_stumps_proba, nfoldCV_stumps,
-       majority_vote, ConfusionMatrix, confusion_matrix, mean_squared_error, R2, load_data,
-       impurity_importance, split_importance, permutation_importance, accuracy
+       load_data, impurity_importance, split_importance, permutation_importance
 
 # ScikitLearn API
 export DecisionTreeClassifier, DecisionTreeRegressor, RandomForestClassifier,

src/classification/main.jl

@@ -194,10 +194,10 @@ Prune tree based on prediction accuracy of each node.
   will be pruned and become a leaf.
 
 * `loss`: The loss function for computing node impurity. Available function include
-  `DecisionTree.util.entropy`, `DecisionTree.util.gini` and `mean_squared_error`. Defaults
-  are `DecisionTree.util.entropy` and `mean_squared_error` for classification tree and
-  regression tree, respectively. If the tree is not a `Root`, this argument does not affect
-  the result.
+  `DecisionTree.util.entropy`, `DecisionTree.util.gini` and
+  `DecisionTree.mean_squared_error`. Defaults are `entropy` and `mean_squared_error` for
+  classification tree and regression tree, respectively. If the tree is not a `Root`, this
+  argument does not affect the result.
 
 For a tree of type `Root`, when any of its nodes is pruned, the `featim` field will be
 updated by recomputing the impurity decrease of that node divided by the total number of

test/runtests.jl

@@ -9,6 +9,9 @@ using Statistics
 using Test
 using LinearAlgebra
 
+import DecisionTree: accuracy, R2, majority_vote, mean_squared_error
+import DecisionTree: confusion_matrix, ConfusionMatrix
+
 println("Julia version: ", VERSION)
 
 similarity(a, b) = first(reshape(a, 1, :) * b / norm(a) / norm(b))