com.cloudera.oryx.app.pmml

Class AppPMMLUtils

java.lang.Object

com.cloudera.oryx.app.pmml.AppPMMLUtils

public final class AppPMMLUtils
extends Object

General app tier PMML-related utility methods.

Method Summary

Modifier and Type	Method and Description
static void	addExtension(org.dmg.pmml.PMML pmml, String key, Object value)
static void	addExtensionContent(org.dmg.pmml.PMML pmml, String key, Collection<?> content)
static CategoricalValueEncodings	buildCategoricalValueEncodings(org.dmg.pmml.DataDictionary dictionary)
static org.dmg.pmml.DataDictionary	buildDataDictionary(InputSchema schema, CategoricalValueEncodings categoricalValueEncodings)
static org.dmg.pmml.MiningSchema	buildMiningSchema(InputSchema schema)
static org.dmg.pmml.MiningSchema	buildMiningSchema(InputSchema schema, double[] importances)
static Integer	findTargetIndex(org.dmg.pmml.MiningSchema miningSchema)
static List<String>	getExtensionContent(org.dmg.pmml.PMML pmml, String name)
static String	getExtensionValue(org.dmg.pmml.PMML pmml, String name)
static List<String>	getFeatureNames(org.dmg.pmml.DataDictionary dictionary)
static List<String>	getFeatureNames(org.dmg.pmml.MiningSchema miningSchema)
static org.dmg.pmml.PMML	readPMMLFromUpdateKeyMessage(String key, String message, org.apache.hadoop.conf.Configuration hadoopConf)
static org.dmg.pmml.Array	toArray(double... values)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

getExtensionValue

public static String getExtensionValue(org.dmg.pmml.PMML pmml,
                                       String name)

getExtensionContent

public static List<String> getExtensionContent(org.dmg.pmml.PMML pmml,
                                               String name)

Parameters:

pmml - PMML model to query for extensions

name - name of extension to query

Returns:

content of the extension, parsed as if it were a PMML Array: space-separated values, with PMML quoting rules

addExtension

public static void addExtension(org.dmg.pmml.PMML pmml,
                                String key,
                                Object value)

Parameters:

pmml - PMML model to add extension to, with no content. It may possibly duplicate existing extensions.

key - extension key

value - extension value

addExtensionContent

public static void addExtensionContent(org.dmg.pmml.PMML pmml,
                                       String key,
                                       Collection<?> content)

Parameters:

pmml - PMML model to add extension to, with a single String content and no value. The content is encoded as if they were being added to a PMML Array and are space-separated with PMML quoting rules

key - extension key

content - list of values to add as a String

toArray

public static org.dmg.pmml.Array toArray(double... values)

Parameters:

values - double value to make into a PMML Array

Returns:

PMML Array representation

buildMiningSchema

public static org.dmg.pmml.MiningSchema buildMiningSchema(InputSchema schema)

Parameters:

schema - InputSchema whose information should be encoded in PMML

Returns:

a MiningSchema representing the information contained in an InputSchema

buildMiningSchema

public static org.dmg.pmml.MiningSchema buildMiningSchema(InputSchema schema,
                                                          double[] importances)

Parameters:

schema - InputSchema whose information should be encoded in PMML

importances - optional feature importances. May be null, or else the size of the array must match the number of predictors in the schema, which may be less than the total number of features.

Returns:

a MiningSchema representing the information contained in an InputSchema

getFeatureNames

public static List<String> getFeatureNames(org.dmg.pmml.MiningSchema miningSchema)

Parameters:

miningSchema - MiningSchema from a model

Returns:

names of features in order

findTargetIndex

public static Integer findTargetIndex(org.dmg.pmml.MiningSchema miningSchema)

Parameters:

miningSchema - MiningSchema from a model

Returns:

index of the MiningField.UsageType#PREDICTED feature

buildDataDictionary

public static org.dmg.pmml.DataDictionary buildDataDictionary(InputSchema schema,
                                                              CategoricalValueEncodings categoricalValueEncodings)

getFeatureNames

public static List<String> getFeatureNames(org.dmg.pmml.DataDictionary dictionary)

Parameters:

dictionary - DataDictionary from model

Returns:

names of features in order

buildCategoricalValueEncodings

public static CategoricalValueEncodings buildCategoricalValueEncodings(org.dmg.pmml.DataDictionary dictionary)

readPMMLFromUpdateKeyMessage

public static org.dmg.pmml.PMML readPMMLFromUpdateKeyMessage(String key,
                                                             String message,
                                                             org.apache.hadoop.conf.Configuration hadoopConf)
                                                      throws IOException

Throws:

IOException