Mailing-List: contact issues-help@flink.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@flink.apache.org
From: greghogan <git@git.apache.org>
To: issues@flink.incubator.apache.org
Reply-To: issues@flink.incubator.apache.org
References: <git-pr-1980-flink@git.apache.org>
In-Reply-To: <git-pr-1980-flink@git.apache.org>
Subject: [GitHub] flink pull request: [FLINK-3780] [gelly] Jaccard Similarity
Content-Type: text/plain
Message-Id: <20160519211613.3BB3BDFDD0@git1-us-west.apache.org>
Date: Thu, 19 May 2016 21:16:13 +0000 (UTC)
archived-at: Thu, 19 May 2016 21:16:18 -0000

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

    https://github.com/apache/flink/pull/1980#discussion_r63956013
  
    --- Diff: flink-libraries/flink-gelly/src/main/java/org/apache/flink/graph/library/similarity/JaccardIndex.java ---
    @@ -0,0 +1,462 @@
    +/*
    + * 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.flink.graph.library.similarity;
    +
    +import org.apache.flink.api.common.ExecutionConfig;
    +import org.apache.flink.api.common.functions.FlatMapFunction;
    +import org.apache.flink.api.common.functions.GroupReduceFunction;
    +import org.apache.flink.api.common.operators.Order;
    +import org.apache.flink.api.java.DataSet;
    +import org.apache.flink.api.java.functions.FunctionAnnotation;
    +import org.apache.flink.api.java.tuple.Tuple2;
    +import org.apache.flink.api.java.tuple.Tuple3;
    +import org.apache.flink.api.java.tuple.Tuple4;
    +import org.apache.flink.graph.Edge;
    +import org.apache.flink.graph.Graph;
    +import org.apache.flink.graph.GraphAlgorithm;
    +import org.apache.flink.graph.asm.degree.annotate.undirected.EdgeTargetDegree;
    +import org.apache.flink.graph.library.similarity.JaccardIndex.Result;
    +import org.apache.flink.graph.utils.Murmur3_32;
    +import org.apache.flink.types.CopyableValue;
    +import org.apache.flink.types.IntValue;
    +import org.apache.flink.types.LongValue;
    +import org.apache.flink.util.Collector;
    +import org.apache.flink.util.Preconditions;
    +
    +import java.util.ArrayList;
    +import java.util.List;
    +
    +/**
    + * The Jaccard Index measures the similarity between vertex neighborhoods.
    + * Scores range from 0.0 (no common neighbors) to 1.0 (all neighbors are common).
    + * <br/>
    + * This implementation produces similarity scores for each pair of vertices
    + * in the graph with at least one common neighbor; equivalently, this is the
    + * set of all non-zero Jaccard Similarity coefficients.
    + * <br/>
    + * The input graph must be a simple, undirected graph containing no duplicate
    + * edges or self-loops.
    + *
    + * @param <K> graph ID type
    + * @param <VV> vertex value type
    + * @param <EV> edge value type
    + */
    +public class JaccardIndex<K extends CopyableValue<K>, VV, EV>
    +implements GraphAlgorithm<K, VV, EV, DataSet<Result<K>>> {
    +
    +	public static final int DEFAULT_GROUP_SIZE = 64;
    +
    +	// Optional configuration
    +	private int groupSize = DEFAULT_GROUP_SIZE;
    +
    +	private boolean unboundedScores = true;
    +
    +	private int minimumScoreNumerator = 0;
    +
    +	private int minimumScoreDenominator = 1;
    +
    +	private int maximumScoreNumerator = 1;
    +
    +	private int maximumScoreDenominator = 0;
    +
    +	private int littleParallelism = ExecutionConfig.PARALLELISM_UNKNOWN;
    +
    +	/**
    +	 * Override the default group size for the quadratic expansion of neighbor
    +	 * pairs. Small groups generate more data whereas large groups distribute
    +	 * computation less evenly among tasks.
    +	 *
    +	 * @param groupSize the group size for the quadratic expansion of neighbor pairs
    +	 * @return this
    +	 */
    +	public JaccardIndex<K, VV, EV> setGroupSize(int groupSize) {
    --- End diff --
    
    I don't know that 64 is the best value but I think it is close. I clarified the javadoc to suggest that users not change this value.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---