Github user greghogan commented on a diff in the pull request:
https://github.com/apache/flink/pull/2885#discussion_r102996700
--- Diff:
flink-examples/flink-examples-batch/src/main/java/org/apache/flink/examples/java/ap/AffinityPropagationBulk.java
---
@@ -0,0 +1,449 @@
+/*
+ * 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.examples.java.ap;
+
+import org.apache.flink.api.common.aggregators.ConvergenceCriterion;
+import org.apache.flink.api.common.aggregators.LongSumAggregator;
+import org.apache.flink.api.common.functions.FilterFunction;
+import org.apache.flink.api.common.functions.GroupReduceFunction;
+import org.apache.flink.api.common.functions.JoinFunction;
+import org.apache.flink.api.common.functions.MapFunction;
+import org.apache.flink.api.common.functions.RichJoinFunction;
+import org.apache.flink.api.common.operators.Order;
+import org.apache.flink.api.java.DataSet;
+import org.apache.flink.api.java.ExecutionEnvironment;
+import org.apache.flink.api.java.operators.IterativeDataSet;
+import org.apache.flink.api.java.tuple.Tuple3;
+import org.apache.flink.api.java.tuple.Tuple4;
+import org.apache.flink.examples.java.ap.util.AffinityPropagationData;
+import org.apache.flink.types.DoubleValue;
+import org.apache.flink.types.LongValue;
+import org.apache.flink.util.Collector;
+import
org.apache.flink.api.java.functions.FunctionAnnotation.ForwardedFieldsFirst;
+import
org.apache.flink.api.java.functions.FunctionAnnotation.ForwardedFieldsSecond;
+import
org.apache.flink.api.java.functions.FunctionAnnotation.ForwardedFields;
+
+/**
+ * Created by joseprubio on 9/22/16.
+ */
+
+public class AffinityPropagationBulk {
+
+ private static final double DAMPING_FACTOR = 0.9;
+ private static final double CONVERGENCE_THRESHOLD = 0.12;
+ private static final String CONVERGENCE_MESSAGES = "message
convergence";
+
+ public static void main(String[] args) throws Exception {
+
+ ExecutionEnvironment env =
ExecutionEnvironment.getExecutionEnvironment();
+ env.getConfig().enableObjectReuse();
+
+ // Get input similarities Tuple3<src, target, similarity>
+ DataSet<Tuple3<LongValue, LongValue, DoubleValue>> similarities
=
+ AffinityPropagationData.getTuplesFromFile(env);
+
+ // Init input to iteration
+ DataSet<Tuple4<LongValue, LongValue, DoubleValue, DoubleValue>>
initMessages
+ = similarities.map(new InitMessage());
+
+ // Iterate
+ IterativeDataSet<Tuple4<LongValue, LongValue, DoubleValue,
DoubleValue>> messages
+ = initMessages.iterate(20);
+
+ // Create aggregator
+
messages.registerAggregationConvergenceCriterion(CONVERGENCE_MESSAGES, new
LongSumAggregator(),
+ new MessageConvergence(similarities.count() * 2));
+
+ // Start responsibility message calculation
+ // r(i,k) <- s(i,k) - max {a(i,K) + s(i,K)} st K != k
+ // Iterate over Tuple6 <Source, Target, Responsibility ,
Availability, IsExemplar, ConvergenceCounter>
+
+ DataSet<Tuple3<LongValue, LongValue, DoubleValue>>
responsibilities = similarities
+
+ // Get a list of a(i,K) + s(i,K) values joining
similarities with messages
+
.join(messages).where("f0","f1").equalTo("f0","f1").with(new
joinAvailabilitySimilarity())
+
+ // Get a dataset with 2 higher values
+ .groupBy("f1").sortGroup("f2",
Order.DESCENDING).first(2)
--- End diff --
Can we reduce on the values of the two highest scores and the vertex ID of
the highest score rather than sorting all group values? This would replace the
following `reduceGroup` as well.
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
