[
https://issues.apache.org/jira/browse/FLINK-2254?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15534249#comment-15534249
]
ASF GitHub Bot commented on FLINK-2254:
---------------------------------------
Github user mushketyk commented on a diff in the pull request:
https://github.com/apache/flink/pull/2564#discussion_r81246840
--- Diff:
flink-libraries/flink-gelly/src/main/java/org/apache/flink/graph/BipartiteGraph.java
---
@@ -0,0 +1,231 @@
+/*
+ * 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;
+
+import org.apache.flink.api.common.functions.FilterFunction;
+import org.apache.flink.api.common.functions.GroupReduceFunction;
+import org.apache.flink.api.java.DataSet;
+import org.apache.flink.api.java.ExecutionEnvironment;
+import org.apache.flink.api.java.functions.KeySelector;
+import org.apache.flink.api.java.tuple.Tuple2;
+
+/**
+ *
+ * Bipartite graph is a graph that contains two sets of vertices: top
vertices and bottom vertices. Edges can only exist
+ * between a pair of vertices from different vertices sets. E.g. there can
be no vertices between a pair
+ * of top vertices.
+ *
+ * <p>Bipartite graph is useful to represent graphs with two sets of
objects, like researchers and their publications,
+ * where an edge represents that a particular publication was authored by
a particular author.
+ *
+ * <p>Bipartite interface is different from {@link Graph} interface, so to
apply algorithms that work on a regular graph
+ * a bipartite graph should be first converted into a {@link Graph}
instance. This can be achieved by using
+ * {@link BipartiteGraph#topProjection(GroupReduceFunction)} or
+ * {@link BipartiteGraph#bottomProjection(GroupReduceFunction)} methods.
+ *
+ * @param <KT> the key type of the top vertices
+ * @param <KB> the key type of the bottom vertices
+ * @param <VT> the top vertices value type
+ * @param <VB> the bottom vertices value type
+ * @param <EV> the edge value type
+ */
+public class BipartiteGraph<KT, KB, VT, VB, EV> {
+ private final ExecutionEnvironment context;
+ private final DataSet<Vertex<KT, VT>> topVertices;
+ private final DataSet<Vertex<KB, VB>> bottomVertices;
+ private final DataSet<BipartiteEdge<KT, KB, EV>> edges;
+
+ private BipartiteGraph(
+ DataSet<Vertex<KT, VT>> topVertices,
+ DataSet<Vertex<KB, VB>> bottomVertices,
+ DataSet<BipartiteEdge<KT, KB, EV>> edges,
+ ExecutionEnvironment context) {
+ this.topVertices = topVertices;
+ this.bottomVertices = bottomVertices;
+ this.edges = edges;
+ this.context = context;
+ }
+
+ /**
+ * Create bipartite graph from datasets.
+ *
+ * @param topVertices dataset of top vertices in the graph
+ * @param bottomVertices dataset of bottom vertices in the graph
+ * @param edges dataset of edges between vertices
+ * @param context Flink execution context
+ * @param <KT> the key type of the top vertices
+ * @param <KB> the key type of the bottom vertices
+ * @param <VT> the top vertices value type
+ * @param <VB> the bottom vertices value type
+ * @param <EV> the edge value type
+ * @return new bipartite graph created from provided datasets
+ */
+ public static <KT, KB, VT, VB, EV> BipartiteGraph<KT, KB, VT, VB, EV>
fromDataSet(
+ DataSet<Vertex<KT, VT>> topVertices,
+ DataSet<Vertex<KB, VB>> bottomVertices,
+ DataSet<BipartiteEdge<KT, KB, EV>> edges,
+ ExecutionEnvironment context) {
+ return new BipartiteGraph<>(topVertices, bottomVertices, edges,
context);
+ }
+
+ /**
+ * Get dataset with top vertices.
+ *
+ * @return dataset with top vertices
+ */
+ public DataSet<Vertex<KT, VT>> getTopVertices() {
+ return topVertices;
+ }
+
+ /**
+ * Get dataset with bottom vertices.
+ *
+ * @return dataset with bottom vertices
+ */
+ public DataSet<Vertex<KB, VB>> getBottomVertices() {
+ return bottomVertices;
+ }
+
+ /**
+ * Get dataset with graph edges.
+ *
+ * @return dataset with graph edges
+ */
+ public DataSet<BipartiteEdge<KT, KB, EV>> getEdges() {
+ return edges;
+ }
+
+ /**
+ * Convert a bipartite into a graph that contains only top vertices. An
edge between two vertices in the new
+ * graph will exist only if the original bipartite graph contains a
bottom vertex they both connected to.
+ *
+ * <p>Caller should provide a function that will create an edge between
two top vertices. This function will receive
--- End diff --
I thought that different datasets would require different algorithms to
consolidate a number of connections into a single edge value. Hence the
callback function. But I think Greg's idea is better.
> Add Bipartite Graph Support for Gelly
> -------------------------------------
>
> Key: FLINK-2254
> URL: https://issues.apache.org/jira/browse/FLINK-2254
> Project: Flink
> Issue Type: New Feature
> Components: Gelly
> Affects Versions: 0.10.0
> Reporter: Andra Lungu
> Assignee: Ivan Mushketyk
> Labels: requires-design-doc
>
> A bipartite graph is a graph for which the set of vertices can be divided
> into two disjoint sets such that each edge having a source vertex in the
> first set, will have a target vertex in the second set. We would like to
> support efficient operations for this type of graphs along with a set of
> metrics(http://jponnela.com/web_documents/twomode.pdf).
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
