Return-Path: X-Original-To: apmail-spark-issues-archive@minotaur.apache.org Delivered-To: apmail-spark-issues-archive@minotaur.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id B265F183E6 for ; Wed, 10 Feb 2016 01:25:18 +0000 (UTC) Received: (qmail 28861 invoked by uid 500); 10 Feb 2016 01:25:18 -0000 Delivered-To: apmail-spark-issues-archive@spark.apache.org Received: (qmail 28744 invoked by uid 500); 10 Feb 2016 01:25:18 -0000 Mailing-List: contact issues-help@spark.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Delivered-To: mailing list issues@spark.apache.org Received: (qmail 28725 invoked by uid 99); 10 Feb 2016 01:25:18 -0000 Received: from arcas.apache.org (HELO arcas) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 10 Feb 2016 01:25:18 +0000 Received: from arcas.apache.org (localhost [127.0.0.1]) by arcas (Postfix) with ESMTP id 36D2B2C1F62 for ; Wed, 10 Feb 2016 01:25:18 +0000 (UTC) Date: Wed, 10 Feb 2016 01:25:18 +0000 (UTC) From: "Joseph K. Bradley (JIRA)" To: issues@spark.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (SPARK-10524) Decision tree binary classification with ordered categorical features: incorrect centroid MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/SPARK-10524?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15140164#comment-15140164 ] Joseph K. Bradley commented on SPARK-10524: ------------------------------------------- Note: I only backported to 1.6 since this may affect the results of user code (though it should improve it in general). > Decision tree binary classification with ordered categorical features: incorrect centroid > ----------------------------------------------------------------------------------------- > > Key: SPARK-10524 > URL: https://issues.apache.org/jira/browse/SPARK-10524 > Project: Spark > Issue Type: Bug > Components: ML, MLlib > Affects Versions: 1.5.0, 1.6.0 > Reporter: Joseph K. Bradley > Assignee: Liang-Chi Hsieh > Fix For: 1.6.1, 2.0.0 > > > In DecisionTree and RandomForest binary classification with ordered categorical features, we order categories' bins based on the hard prediction, but we should use the soft prediction. > Here are the 2 places in mllib and ml: > * [https://github.com/apache/spark/blob/45de518742446ddfbd4816c9d0f8501139f9bc2d/mllib/src/main/scala/org/apache/spark/mllib/tree/DecisionTree.scala#L887] > * [https://github.com/apache/spark/blob/45de518742446ddfbd4816c9d0f8501139f9bc2d/mllib/src/main/scala/org/apache/spark/ml/tree/impl/RandomForest.scala#L779] > The PR which fixes this should include a unit test which isolates this issue, ideally by directly calling binsToBestSplit. -- This message was sent by Atlassian JIRA (v6.3.4#6332) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org For additional commands, e-mail: issues-help@spark.apache.org