[ 
https://issues.apache.org/jira/browse/SPARK-23409?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Joseph K. Bradley resolved SPARK-23409.
---------------------------------------
    Resolution: Duplicate

Linking old JIRA for this issue

> RandomForest/DecisionTree (syntactic) pruning of redundant subtrees
> -------------------------------------------------------------------
>
>                 Key: SPARK-23409
>                 URL: https://issues.apache.org/jira/browse/SPARK-23409
>             Project: Spark
>          Issue Type: Improvement
>          Components: MLlib
>    Affects Versions: 2.2.1
>         Environment: 
>            Reporter: Alessandro Solimando
>            Priority: Minor
>
> Improvement: redundancy elimination from decision trees where all the leaves 
> of a given subtree share the same prediction.
> Benefits:
>  * Model interpretability
>  * Faster unitary model invocation (relevant for massive number of 
> invocations)
>  * Smaller model memory footprint
> For instance, consider the following decision tree.
> {panel:title=Original Decision Tree}
> {noformat}
> DecisionTreeClassificationModel (uid=dtc_e794a5a3aa9e) of depth 3 with 15 
> nodes
>   If (feature 1 <= 0.5)
>    If (feature 2 <= 0.5)
>     If (feature 0 <= 0.5)
>      Predict: 0.0
>     Else (feature 0 > 0.5)
>      Predict: 0.0
>    Else (feature 2 > 0.5)
>     If (feature 0 <= 0.5)
>      Predict: 0.0
>     Else (feature 0 > 0.5)
>      Predict: 0.0
>   Else (feature 1 > 0.5)
>    If (feature 2 <= 0.5)
>     If (feature 0 <= 0.5)
>      Predict: 1.0
>     Else (feature 0 > 0.5)
>      Predict: 1.0
>    Else (feature 2 > 0.5)
>     If (feature 0 <= 0.5)
>      Predict: 0.0
>     Else (feature 0 > 0.5)
>      Predict: 0.0
> {noformat}
> {panel}
> The proposed method, taken as input the first tree, aims at producing as 
> output the following (semantically equivalent) tree:
> {panel:title=Pruned Decision Tree}
> {noformat}
> DecisionTreeClassificationModel (uid=dtc_e794a5a3aa9e) of depth 3 with 15 
> nodes
>   If (feature 1 <= 0.5)
>    Predict: 0.0
>   Else (feature 1 > 0.5)
>    If (feature 2 <= 0.5)
>     Predict: 1.0
>    Else (feature 2 > 0.5)
>     Predict: 0.0
> {noformat}
> {panel}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to