com.kogalur.randomforest

Class RandomForestModel

java.lang.Object

com.kogalur.randomforest.RandomForestModel

public class RandomForestModel
extends Object

Class that represents a random forest model as a result of growing, restoring, or predicting using new data.

Author:

Udaya Kogalur

Method Summary

Modifier and Type	Method and Description
int	get_pruningCount() Returns the value representing the number of terminal nodes to which each tree in the original model should be pruned back.
int	get_rfCores() Returns the number of cores to be used by the algorithm when OpenMP parallel processing is in force.
int	get_seed() Returns the seed for the random number generator used by the algorithm.
int	get_trace() Returns the trace parameter indicating the specified update interval in seconds.
String[]	get_xImportance() Returns the x-variable(s) over which importance calculations should be conducted.
String[]	get_yTarget() Returns the array of y-variable to be targeted when multivariate families are in force.
Object	getEnsemble(String name)
String	getEnsembleArg(String key) Returns the current value for the specified ensemble argument.
ModelArg	getModelArg() Returns the class containing the model arguments that produced the RandomForestModel object.
com.kogalur.randomforest.ModelType	getModelType()
void	printEnsemble(Object ensb)
void	printEnsembleList()
void	set_pruningCount(int pruningCount) Sets the number of terminal nodes to which each tree in the original model should be pruned back.
void	set_rfCores(int rfCores) Sets the number of cores to be used by the algorithm when OpenMP parallel processing is in force.
void	set_seed(int seed) Sets the seed for the random number generator used by the algorithm.
void	set_trace(int trace) Sets the trace parameter indicating the specified update interval in seconds.
void	set_xImportance() Sets the x-variable(s), over which importance calculations should be conducted, to all x-variables.
void	set_xImportance(String[] xImportance) Sets the x-variable(s) over which importance calculations should be conducted.
void	set_yTarget() Sets the default action for the y-variable(s) to be targeted when multivariate families are in force.
void	set_yTarget(String[] yTarget) Sets the y-variable(s) to be targeted when multivariate families are in force.
void	setEnsembleArg() Sets default values for the ensemble outputs resulting from restoring the model or using it to predict with new data.
void	setEnsembleArg(String key, String value) Sets the ensemble outputs desired when restoring the model or using it to predict with new data.

Methods inherited from class java.lang.Object

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

Method Detail

getModelArg

public ModelArg getModelArg()

Returns the class containing the model arguments that produced the RandomForestModel object.

Returns:

The model arguments that produced the RandomForestModel object.

getEnsemble

public Object getEnsemble(String name)

setEnsembleArg

public void setEnsembleArg(String key,
                           String value)

Sets the ensemble outputs desired when restoring the model or using it to predict with new data. These outputs are similar to those available when initially creating a model using ModelArg.setEnsembleArg(String, String), but are context sensitive to restoration versus prediction. These settings are in the form of <key, value> pairs, where the key is the name of the ensemble, and the value is the specific option for that ensemble.

The default option for each key is in bold.

  
 

  

    
Ensemble Key
    
Possible Values
  
  
      
  

    
weight
    
no, inbag, oob, all
  

  

    
proximity
    
no, inbag, oob, all
  
  
  

    
membership
    
no, yes
  

  

    
importance
    
no, permute, random, anti, 
 permute.joint, random.joint, anti.joint 
 permute.ensemble, random.ensemble, anti.ensemble, 
 permute.joint.ensemble, random.joint.ensemble, anti.joint.ensemble 
  

  

    
varUsed
    
no, every.tree, sum.tree
  

  

    
splitDepth
    
no, every.tree, sum.tree
  

  

    
errorType
    
For RF-C, RF-C+ Families Only:
default, brier, g.mean, g.mean.drc
  

 
  

Note that the options inbag, oob are only relevant when restoring a model, and all is the only relevant option when predicting with new data. Invalid values will be overridden.

Parameters:

key - The name of the ensemble output.

value - The specific value for the ensemble output.

setEnsembleArg

public void setEnsembleArg()

Sets default values for the ensemble outputs resulting from restoring the model or using it to predict with new data.

See Also:

setEnsembleArg(String, String).

getEnsembleArg

public String getEnsembleArg(String key)

Returns the current value for the specified ensemble argument.

Parameters:

key - The name of the ensemble output.

See Also:

setEnsembleArg(String, String).

set_pruningCount

public void set_pruningCount(int pruningCount)

Sets the number of terminal nodes to which each tree in the original model should be pruned back. The terminal node membership for the pruned forest is returned in the ensemble pstnMembership.

Parameters:

pruningCount - The number of terminal nodes to which each tree in the original model should be pruned back. The default action is to conduct no pruning, and is explicitly achieved by setting this parameter to zero (0).

get_pruningCount

public int get_pruningCount()

Returns the value representing the number of terminal nodes to which each tree in the original model should be pruned back.

See Also:

set_pruningCount(int)

set_yTarget

public void set_yTarget(String[] yTarget)

Sets the y-variable(s) to be targeted when multivariate families are in force. Predictions over this set of targets will be included in the ensembles.

Parameters:

yTarget - Array of y-variables to be targeted when multivariate families are in force. The default action is to use all y-variables.

set_yTarget

public void set_yTarget()

Sets the default action for the y-variable(s) to be targeted when multivariate families are in force. The default action is to use all y-variables.

See Also:

set_yTarget(String[])

get_yTarget

public String[] get_yTarget()

Returns the array of y-variable to be targeted when multivariate families are in force.

See Also:

set_yTarget(String[])

set_xImportance

public void set_xImportance(String[] xImportance)

Sets the x-variable(s) over which importance calculations should be conducted. This is useful when xSize is large and importance over a small subset of x-variables is desired. In addition, if a joint importance calculation is desired, the specified x-variables are jointly targeted in the calculations.

Parameters:

xImportance - Array of x-variables over which important calculations should be conducted. The default action is to not provide importance ensembles. This can be alse be accomplished by sending a null value into the method.

set_xImportance

public void set_xImportance()

Sets the x-variable(s), over which importance calculations should be conducted, to all x-variables. Be cautious when xSize is large, as computation times can be lengthy.

See Also:

set_xImportance(String[])

get_xImportance

public String[] get_xImportance()

Returns the x-variable(s) over which importance calculations should be conducted.

See Also:

set_xImportance(String[])

set_seed

public void set_seed(int seed)

Sets the seed for the random number generator used by the algorithm. This must be a negative number. The seed is a very important parameter if repeatability of the model generated is required. Generally speaking, growing a model using the same data set, the same model paramaters, and the same seed will result in identical models. When large amounts of missing data are involved, there can be slight variations due to Monte Carlo effects. If the parameter is not set by the user, it can always be recovered with get_seed().

get_seed

public int get_seed()

Returns the seed for the random number generator used by the algorithm.

Returns:

The seed for the random number generator used by the

See Also:

set_seed(int)

set_trace

public void set_trace(int trace)

Sets the trace parameter indicating the specified update interval in seconds.

Parameters:

trace - The trace parameter indicating the specified update interval in seconds. During extended execution times, the approximate time to complete the execution is output to a trace file in the users HOME directory. The format and location of the trace file can be controlled by modifying src/main/resources/spark/log.properties. A value of zero (0) turns off the trace.

get_trace

public int get_trace()

Returns the trace parameter indicating the specified update interval in seconds.

Returns:

The trace parameter indicating the specified update interval in seconds.

set_rfCores

public void set_rfCores(int rfCores)

Sets the number of cores to be used by the algorithm when OpenMP parallel processing is in force.

Parameters:

rfCores - The number of cores to be used by the algorithm when OpenMP parallel processing is in force. The default behaviour is to use all cores available. This is achieved by setting the parameter to a negative value. The result is that each core will be independently tasked with growing a tree. Significant savings in elapsed computation times can be achieved.

get_rfCores

public int get_rfCores()

Returns the number of cores to be used by the algorithm when OpenMP parallel processing is in force.

Returns:

the number of cores to be used by the algorithm when OpenMP parallel processing is in force.

See Also:

set_rfCores(int)

getModelType

public com.kogalur.randomforest.ModelType getModelType()

printEnsembleList

public void printEnsembleList()

printEnsemble

public void printEnsemble(Object ensb)