org.apache.lens.ml.impl

Class MLTask

java.lang.Object

org.apache.lens.ml.impl.MLTask

All Implemented Interfaces:

Runnable

public class MLTask
extends Object
implements Runnable

Run a complete cycle of train and test (evaluation) for an ML algorithm

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	MLTask.Builder Builder to create an example task
static class	MLTask.State

Method Summary

Methods

Modifier and Type	Method and Description
String	getAlgorithm() Name of the algo/algorithm.
org.apache.hadoop.hive.conf.HiveConf	getConfiguration() Configuration for the example.
Map<String,String>	getExtraParams() Extra params passed to the training algorithm
List<String>	getFeatureColumns() Names of columns which are features in the training data.
String	getLabelColumn() Name of the column which is a label for supervised algorithms.
LensMLClient	getMlClient() ml client
String	getModelID()
String	getOutputTable() Output table name
String	getPartitionSpec() Training table partition spec
String	getReportID()
MLTask.State	getTaskState()
String	getTestTable() Name of the table containing test data.
String	getTrainingTable() Name of the table containing training data.
void	run()
String	toString()

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, wait, wait, wait

Method Detail

run

public void run()

Specified by:

run in interface Runnable

toString

@Generated(value="lombok")
public String toString()

Overrides:

toString in class Object

getTaskState

@Generated(value="lombok")
public MLTask.State getTaskState()

getAlgorithm

@Generated(value="lombok")
public String getAlgorithm()

Name of the algo/algorithm.

getTrainingTable

@Generated(value="lombok")
public String getTrainingTable()

Name of the table containing training data.

getTestTable

@Generated(value="lombok")
public String getTestTable()

Name of the table containing test data. Optional, if not provided trainingTable itself is used for testing

getPartitionSpec

@Generated(value="lombok")
public String getPartitionSpec()

Training table partition spec

getLabelColumn

@Generated(value="lombok")
public String getLabelColumn()

Name of the column which is a label for supervised algorithms.

getFeatureColumns

@Generated(value="lombok")
public List<String> getFeatureColumns()

Names of columns which are features in the training data.

getConfiguration

@Generated(value="lombok")
public org.apache.hadoop.hive.conf.HiveConf getConfiguration()

Configuration for the example.

getMlClient

@Generated(value="lombok")
public LensMLClient getMlClient()

ml client

getOutputTable

@Generated(value="lombok")
public String getOutputTable()

Output table name

getExtraParams

@Generated(value="lombok")
public Map<String,String> getExtraParams()

Extra params passed to the training algorithm

getModelID

@Generated(value="lombok")
public String getModelID()

getReportID

@Generated(value="lombok")
public String getReportID()