Mailing-List: contact issues-help@mahout.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@mahout.apache.org
Date: Wed, 30 Aug 2017 20:27:00 +0000 (UTC)
From: "ASF GitHub Bot (JIRA)" <jira@apache.org>
To: issues@mahout.apache.org
Message-ID: <JIRA.13077596.1496738089000.163667.1504124820833@Atlassian.JIRA>
In-Reply-To: <JIRA.13077596.1496738089000@Atlassian.JIRA>
References: <JIRA.13077596.1496738089000@Atlassian.JIRA> <JIRA.13077596.1496738089377@jira-lw-us.apache.org>
Subject: [jira] [Commented] (MAHOUT-1991) Implement naive DBSCAN Algorithm -
 O(n^2) complexity
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Wed, 30 Aug 2017 20:27:08 -0000


    [ https://issues.apache.org/jira/browse/MAHOUT-1991?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16147995#comment-16147995 ] 

ASF GitHub Bot commented on MAHOUT-1991:
----------------------------------------

Github user andrewpalumbo commented on a diff in the pull request:

    https://github.com/apache/mahout/pull/334#discussion_r136179872
  
    --- Diff: math-scala/src/main/scala/org/apache/mahout/math/algorithms/clustering/DBSCAN.scala ---
    @@ -0,0 +1,217 @@
    +/**
    +  * Licensed to the Apache Software Foundation (ASF) under one
    +  * or more contributor license agreements. See the NOTICE file
    +  * distributed with this work for additional information
    +  * regarding copyright ownership. The ASF licenses this file
    +  * to you under the Apache License, Version 2.0 (the
    +  * "License"); you may not use this file except in compliance
    +  * with the License. You may obtain a copy of the License at
    +  *
    +  * http://www.apache.org/licenses/LICENSE-2.0
    +  *
    +  * Unless required by applicable law or agreed to in writing,
    +  * software distributed under the License is distributed on an
    +  * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
    +  * KIND, either express or implied. See the License for the
    +  * specific language governing permissions and limitations
    +  * under the License.
    +  */
    +
    +package org.apache.mahout.math.algorithms.clustering
    +
    +import org.apache.mahout.math.scalabindings._
    +import org.apache.mahout.math.scalabindings.RLikeOps._
    +import org.apache.mahout.math._
    +import org.apache.mahout.math.algorithms.common.distance.{DistanceMetric, DistanceMetricSelector}
    +import org.apache.mahout.math.drm._
    +import org.apache.mahout.math.drm.RLikeDrmOps._
    +
    +import scala.collection.mutable
    +import scala.io.Source
    +
    +class DistributedDBSCAN extends ClusteringFitter {
    +
    +  var epsilon: Double = _
    +  var minPts: Int = _
    +  var distanceMeasure: Symbol = _
    +
    +  def setStandardHyperparameters(hyperparameters: Map[Symbol, Any] = Map('foo -> None)): Unit = {
    +    epsilon = hyperparameters.asInstanceOf[Map[Symbol, Double]].getOrElse('epsilon, 0.5)
    +    minPts = hyperparameters.asInstanceOf[Map[Symbol, Int]].getOrElse('minPts, 1)
    +    distanceMeasure = hyperparameters.asInstanceOf[Map[Symbol, Symbol]].getOrElse('distanceMeasure, 'Euclidean)
    +  }
    +
    +  def fit[K](input: DrmLike[K],
    +             hyperparameters: (Symbol, Any)*): DBSCANModel = {
    +
    +    setStandardHyperparameters(hyperparameters.toMap)
    +    implicit val ctx = input.context
    +    implicit val ktag =  input.keyClassTag
    +
    +    val dmNumber = DistanceMetricSelector.namedMetricLookup(distanceMeasure)
    +
    +    val configBC = drmBroadcast(dvec(epsilon, minPts, dmNumber))
    +
    +    val clusters = input.allreduceBlock(
    +      {
    +        // Assign All Points to Clusters
    +        case (keys, block: Matrix) => {
    +          val epsilon_local = configBC.value.get(0)
    +          val minPts_local = configBC.value.get(1)
    +
    +          val distanceMetric = DistanceMetricSelector.select(configBC.value.get(3))
    +          val icDBSCAN = new InCoreDBSCAN(block, epsilon_local, minPts_local.toInt, distanceMetric)
    +          // do stuff on icDBSCAN
    +          icDBSCAN.DBSCAN()
    +        }
    +      }, {
    +        // Optionally Merge Clusters that are close enough
    +        case (metadata1: Matrix, metadata2: Matrix) => {
    +          // this does nothing- just returns the left matrix
    +          metadata1
    +        }
    +      })
    +
    +    val model = new DBSCANModel(1)
    +    model.summary = s"""foo the bar"""
    --- End diff --
    
    @AdityaAS do you have a plan for the model summary?


> Implement naive DBSCAN Algorithm - O(n^2) complexity
> ----------------------------------------------------
>
>                 Key: MAHOUT-1991
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-1991
>             Project: Mahout
>          Issue Type: New Feature
>          Components: Algorithms
>            Reporter: Aditya AS
>            Assignee: Aditya AS
>
> Implement the naive DBSCAN algorithm in Mahout Samsara, as part of the Algorithms Framework.


--
This message was sent by Atlassian JIRA
(v6.4.14#64029)