magpie.models.regression

Class ScikitLearnRegression

java.lang.Object

magpie.models.BaseModel

magpie.models.regression.BaseRegression

magpie.models.regression.ScikitLearnRegression

All Implemented Interfaces:

java.io.Serializable, java.lang.AutoCloseable, java.lang.Cloneable, ExternalModel, AbstractRegressionModel, Citable, Commandable, Options, Printable, Savable

public class ScikitLearnRegression
extends BaseRegression
implements ExternalModel

Uses Scikit-learn to train a regression model. User must provide the path to a pickle file containing a Python object that fulfills two operations:

1. fit(X, y): Train the model given attribute matrix X, and observation matrix y.
2. predict(X): Run the model

Note: This implementation requires the ability to write temporary files on whatever system Magpie is running on.

Usage: <model> [<compression level>]
model: Path to a file containing a Scikit-Learn regression object
compression: Degree to which model is compressed before storing in Magpie. 0: fastest, 9: lowest memory footprint (default: 5)
See this tutorial for how to save Scikit-Learn objects.

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
boolean	Debug Whether to print output from python stdout to screen
protected java.lang.Process	ScikitServer Scikit server process

Fields inherited from class magpie.models.BaseModel

AttributeSelector, trained, TrainingStats, validated, ValidationStats

Constructor Summary

Constructors

Constructor and Description
ScikitLearnRegression()

Method Summary

Modifier and Type	Method and Description
ScikitLearnRegression	clone()
void	close()
void	closeServer() Set a command to the model server that we are done with it.
int	getNFittingParameters() Number of fitting parameters in a model.
int	getPort() Get the port number for this model's server
java.lang.Process	getProcess() Get the process holding this external port
protected java.lang.String	printModel_protected() Internal method that handles printing the model as a string.
java.util.List<java.lang.String>	printModelDescriptionDetails(boolean htmlFormat) Print details of the model.
java.lang.String	printUsage() Print out required format for options.
void	readModel(java.io.InputStream input) Read model from an input stream
void	run_protected(Dataset runData) Run a model without checking if stuff is trained (use carefully)
boolean	serverIsRunning() Check whether the server is running
void	setCompressionLevel(int level) Define how well model is compressed after training.
void	setOptions(java.util.List<java.lang.Object> Options) Set any options for this object.
void	startServer() Start the server hosting the model.
protected void	train_protected(Dataset TrainData) Train a model without evaluating performance
void	writeModel(java.io.OutputStream output) Write model to output stream

Methods inherited from class magpie.models.regression.BaseRegression

doRobustRegression, getRobustRegressionQ, robustTraining, runCommand, setRobustRegressionQ

Methods inherited from class magpie.models.BaseModel

about, crossValidate, crossValidate, crossValidate, done, externallyValidate, getAttributeSelector, getCitations, getFilter, getTrainTime, getValidationMethod, handleSetCommand, isTrained, isValidated, loadState, printCommand, printDescription, printModel, resetModel, run, saveCommand, saveState, setAttributeSelector, setComponent, setFilter, train, train

Methods inherited from class java.lang.Object

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

Field Detail

ScikitServer

protected transient java.lang.Process ScikitServer

Scikit server process

Debug

public boolean Debug

Whether to print output from python stdout to screen

Constructor Detail

ScikitLearnRegression

public ScikitLearnRegression()

Method Detail

close

public void close()
           throws java.lang.Exception

Specified by:

close in interface java.lang.AutoCloseable

Throws:

java.lang.Exception

closeServer

public void closeServer()

Description copied from interface: ExternalModel

Set a command to the model server that we are done with it. Use when program is done with a model, and needs to clear up port

Specified by:

closeServer in interface ExternalModel

clone

public ScikitLearnRegression clone()

Overrides:

clone in class BaseRegression

setOptions

public void setOptions(java.util.List<java.lang.Object> Options)
                throws java.lang.Exception

Description copied from interface: Options

Set any options for this object. If parsing the options fails, return an Exception with the usage as its message.

Specified by:

setOptions in interface Options

Parameters:

Options - Array of options as Objects - can be null

Throws:

java.lang.Exception - if problem with inputs

printUsage

public java.lang.String printUsage()

Description copied from interface: Options

Print out required format for options. For consistency, should start with "Usage: ".

Specified by:

printUsage in interface Options

Returns:

Usage for this function

getPort

public int getPort()

Description copied from interface: ExternalModel

Get the port number for this model's server

Specified by:

getPort in interface ExternalModel

Returns:

Port number. If server is not running, this is meaningless

setCompressionLevel

public void setCompressionLevel(int level)
                         throws java.lang.Exception

Define how well model is compressed after training.

This class works by launching a server than runs a scikit-learn model. After training, this server sends back the model as a pickle file. For large datasets, this could be a huge file. This option allows one to compress it before transmission.

Specified by:

setCompressionLevel in interface ExternalModel

Parameters:

level - Desired level. 1: Fastest, 9: Smallest memory footprint

Throws:

java.lang.Exception

See Also:

ScikitModel

readModel

public void readModel(java.io.InputStream input)
               throws java.lang.Exception

Read model from an input stream

Specified by:

readModel in interface ExternalModel

Parameters:

input - Input stream providing model data

Throws:

java.io.IOException

java.lang.Exception

writeModel

public void writeModel(java.io.OutputStream output)
                throws java.io.IOException

Write model to output stream

Specified by:

writeModel in interface ExternalModel

Parameters:

output -

Throws:

java.io.IOException

startServer

public void startServer()
                 throws java.lang.Exception

Description copied from interface: ExternalModel

Start the server hosting the model. This process will initialize the server and set the communication port number.

Specified by:

startServer in interface ExternalModel

Throws:

java.lang.Exception

serverIsRunning

public boolean serverIsRunning()

Description copied from interface: ExternalModel

Check whether the server is running

Specified by:

serverIsRunning in interface ExternalModel

Returns:

Whether server is running

getProcess

public java.lang.Process getProcess()

Description copied from interface: ExternalModel

Get the process holding this external port

Specified by:

getProcess in interface ExternalModel

Returns:

train_protected

protected void train_protected(Dataset TrainData)

Description copied from class: BaseModel

Train a model without evaluating performance

Specified by:

train_protected in class BaseModel

Parameters:

TrainData - Training data

run_protected

public void run_protected(Dataset runData)

Description copied from class: BaseModel

Run a model without checking if stuff is trained (use carefully)

Specified by:

run_protected in class BaseModel

Parameters:

runData - Training data

printModel_protected

protected java.lang.String printModel_protected()

Description copied from class: BaseModel

Internal method that handles printing the model as a string. Should end with a "\n"

Specified by:

printModel_protected in class BaseModel

Returns:

String representation of model

getNFittingParameters

public int getNFittingParameters()

Description copied from interface: AbstractRegressionModel

Number of fitting parameters in a model. Only pertinent to nonlinear regression. Otherwise, this should just return 0.

Specified by:

getNFittingParameters in interface AbstractRegressionModel

Returns:

Number of fitting parameters

printModelDescriptionDetails

public java.util.List<java.lang.String> printModelDescriptionDetails(boolean htmlFormat)

Description copied from class: BaseModel

Print details of the model. Used by BaseModel.printDescription(boolean).

Implementation note: No not add indentation for details. That is handled by BaseModel.printDescription(boolean). You should also call the super operation to get the Normalizer and Attribute selector settings

Overrides:

printModelDescriptionDetails in class BaseModel

Parameters:

htmlFormat - Whether to use HTML format

Returns:

List describing model details. Each entry is a different line of the description (i.e., in place of newline characters).