Package | Description |
---|---|
jsat.classifiers.trees |
Modifier and Type | Class and Description |
---|---|
class |
ImportanceByUses |
class |
MDA
Mean Decrease in Accuracy (MDA) measures feature importance by applying the
classifier for each feature, and corruption one feature at a time as each
dataum its pushed through the tree.
|
class |
MDI
Determines the importance of features by measuring the decrease in impurity
caused by each feature used, weighted by the amount of data seen by the node
using the feature.
|
Modifier and Type | Method and Description |
---|---|
<Type extends DataSet> |
ERTrees.evaluateFeatureImportance(DataSet<Type> data,
TreeFeatureImportanceInference imp)
Measures the statistics of feature importance from the trees in this
forest.
|
Copyright © 2017. All rights reserved.