Introduction
As explained in Feature Management in WSO2 Admin Guide - Working with Features , each WSO2 product is a collection of reusable software units called features. A single feature is a list of components and/or other features. This section describes how to install WSO2 Machine Learner features in WSO2 CEP.
...
| Info |
|---|
| Note |
|---|
When installing ML features in an Apache Storm cluster, it is recommended to use a pom file instead of the Management Console. The Management Control only allows the features to be installed in the default profile instead of in all the profiles of the CEP nodes, and as a result, an exception occurs when events are sent to nodes that do not use the default profile. For more information, see Installing Features using pom Files. |
When you run WSO2 CEP in a distributed mode, the following needs to be carried out <CEP_HOME>/samples/cep/utils/storm-dependencies.jar/pom.xml The following dependencies should be uncommented in the <CEP_HOME>/samples/cep/utils/storm-dependencies.jar/pom.xml file as shown below.
| Code Block |
|---|
| <!-- Uncomment the following depedency section if you want to include Siddhi ML extension as part of
Storm dependencies -->
<dependency>
<groupId>org.wso2.carbon.ml</groupId>
<artifactId>org.wso2.carbon.ml.siddhi.extension</artifactId>
<version>${carbon.ml.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.carbon.ml</groupId>
<artifactId>org.wso2.carbon.ml.core</artifactId>
<version>${carbon.ml.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.carbon.ml</groupId>
<artifactId>org.wso2.carbon.ml.database</artifactId>
<version>${carbon.ml.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.carbon.ml</groupId>
<artifactId>org.wso2.carbon.ml.commons</artifactId>
<version>${carbon.ml.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.carbon.metrics</groupId>
<artifactId>org.wso2.carbon.metrics.manager</artifactId>
<version>${carbon.metrics.version}</version>
</dependency>
<!--<!– Dependencies for Spark –>-->
<dependency>
<groupId>org.wso2.orbit.org.apache.spark</groupId>
<artifactId>spark-core_2.10</artifactId>
<version>${spark.core.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.orbit.org.apache.spark</groupId>
<artifactId>spark-sql_2.10</artifactId>
<version>${spark.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.orbit.org.apache.spark</groupId>
<artifactId>spark-mllib_2.10</artifactId>
<version>${spark.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.orbit.org.apache.spark</groupId>
<artifactId>spark-streaming_2.10</artifactId>
<version>${spark.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.orbit.org.scalanlp</groupId>
<artifactId>breeze_2.10</artifactId>
<version>${breeze.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.orbit.jblas</groupId>
<artifactId>jblas</artifactId>
<version>${jblas.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.orbit.spire-math</groupId>
<artifactId>spire_2.10</artifactId>
<version>${spire.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.orbit.org.apache.hadoop</groupId>
<artifactId>hadoop-client</artifactId>
<version>${hadoop.client.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.uncommons.maths</groupId>
<artifactId>uncommons-maths</artifactId>
<version>${uncommons.maths.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.json4s</groupId>
<artifactId>json4s-jackson_2.10</artifactId>
<version>${json4s.jackson.version}</version>
</dependency>
<dependency>
<groupId>org.slf4j</groupId>
<artifactId>slf4j-api</artifactId>
<version>${slf4j.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.orbit.github.fommil.netlib</groupId>
<artifactId>core</artifactId>
<version>${fommil.netlib.version}</version>
</dependency>
<dependency>
<groupId>org.wso2.orbit.sourceforge.f2j</groupId>
<artifactId>arpack_combined</artifactId>
<version>${arpack.combined.version}</version>
</dependency>
<dependency>
<groupId>org.scala-lang</groupId>
<artifactId>scala-library</artifactId>
<version>${scala.version}</version>
</dependency>
<dependency>
<groupId>org.apache.commons</groupId>
<artifactId>commons-csv</artifactId>
<version>${commons.csv.version}</version>
</dependency> |
| Code Block |
|---|
| <!-- ML extension dependencies -->
<include>org.wso2.orbit.org.apache.spark:spark-core_2.10
</include>
<include>org.wso2.orbit.org.apache.spark:spark-sql_2.10
</include>
<include>org.wso2.orbit.org.apache.spark:spark-mllib_2.10
</include>
<include>org.wso2.orbit.org.apache.spark:spark-streaming_2.10
</include>
<include>org.wso2.orbit.org.scalanlp:breeze_2.10</include>
<include>org.wso2.orbit.jblas:jblas</include>
<include>org.wso2.orbit.spire-math:spire_2.10</include>
<include>org.wso2.orbit.org.apache.hadoop:hadoop-client
</include>
<include>org.wso2.uncommons.maths:uncommons-maths</include>
<include>org.wso2.json4s:json4s-jackson_2.10</include>
<include>org.slf4j:slf4j-api</include>
<include>org.wso2.orbit.github.fommil.netlib:core</include>
<include>org.wso2.orbit.sourceforge.f2j:arpack_combined
</include>
<include>org.scala-lang:scala-library</include>
<include>org.apache.commons:commons-csv</include>
<include>org.wso2.carbon.ml:org.wso2.carbon.ml.core</include>
<include>org.wso2.carbon.ml:org.wso2.carbon.ml.database
</include>
<include>org.wso2.carbon.ml:org.wso2.carbon.ml.commons</include>
<include>
org.wso2.carbon.ml:org.wso2.carbon.ml.siddhi.extension
</include>
<include>
org.wso2.carbon.metrics:org.wso2.carbon.metrics.manager
</include> |
Run the following command from the <CEP_HOME>/samples/cep/utils/storm-dependencies-jar directory.
mvn clean install
This will generate a jar in the target directory.
| Note |
|---|
Queries are collectively processed by all the nodes in a CEP cluster. Therefore, make sure that the ML model is located in the same path in all the nodes. This allows all the nodes to access the model when events are sent to a specific node. |
|