FAQs - Revision history

Grau at 15:04, 2 February 2012

2012-02-02T15:04:05Z

← Older revision		Revision as of 15:04, 2 February 2012
Line 71:		Line 71:
	== Implementing new models ==		== Implementing new models ==
	<p>		<p>
	'''Q: How do I implement a new generative [http://www.jstacs.de/api/de/jstacs/~~models~~/~~Model~~.html ~~Model~~]?'''<br />		'''Q: How do I implement a new generative [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html TrainableStatisticalModel]?'''<br />
	'''A:''' Write an implementation of the [http://www.jstacs.de/api/de/jstacs/~~models~~/~~Model~~.html ~~Model~~] interface. For convenience, you can use the abstract [http://www.jstacs.de/api/de/jstacs/~~models~~/~~AbstractModel~~.html ~~AbstractModel~~] class with default implementations for many methods~~.<br />~~		'''A:''' Write an implementation of the [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html TrainableStatisticalModel] interface. For convenience, you can use the abstract [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/AbstractTrainableStatisticalModel.html AbstractTrainableStatisticalModel] class with default implementations for many methods.
	~~Also see the [[Implementation of a homogeneous Markov model of order 0 based on AbstractModel \| code example]]~~.
	</p>		</p>
	<hr />		<hr />
	<p>		<p>
	'''Q: How do I implement a new discriminative model?'''<br />		'''Q: How do I implement a new discriminative model?'''<br />
	'''A:''' Write an implementation of the [http://www.jstacs.de/api/de/jstacs/~~scoringFunctions~~/~~ScoringFunction~~.html ~~ScoringFunction~~] interface. For convenience, you can use the abstract [http://www.jstacs.de/api/de/jstacs/~~scoringFunctions~~/~~AbstractNormalizableScoringFunction~~.html ~~AbstractNormalizableScoringFunction~~] class with default implementations for many methods.		'''A:''' Write an implementation of the [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/differentiable/DifferentiableStatisticalModel.html DifferentiableStatisticalModel] interface. For convenience, you can use the abstract [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/differentiable/AbstractDifferentiableStatisticalModel.html AbstractDifferentiableStatisticalModel] class with default implementations for many methods.
	</p>		</p>
	<hr />		<hr />
	<p>		<p>
	'''Q: How do I implement a model that can be trained generatively ''and'' discriminatively?'''<br />		'''Q: How do I implement a model that can be trained generatively ''and'' discriminatively?'''<br />
	'''A:''' You can either extend [http://www.jstacs.de/api/de/jstacs/~~models~~/~~AbstractModel~~.html ~~AbstractModel~~] and additionally implement the [http://www.jstacs.de/api/de/jstacs/~~scoringFunctions~~/~~ScoringFunction~~.html ~~ScoringFunction~~] interface, or you extend the [http://www.jstacs.de/api/de/jstacs/~~scoringFunctions~~/~~AbstractNormalizableScoringFunction~~.html ~~AbstractNormalizableScoringFunction~~] and additionally implement the [http://www.jstacs.de/api/de/jstacs/~~models~~/~~Model~~.html ~~Model~~] interface.		'''A:''' You can either extend [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/AbstractTrainableStatisticalModel.html AbstractTrainableStatisticalModel] and additionally implement the [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/differentiable/DifferentiableStatisticalModel.html DifferentiableStatisticalModel] interface, or you extend the [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/differentiable/AbstractDifferentiableStatisticalModel.html AbstractDifferentiableStatisticalModel] and additionally implement the [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html TrainableStatisticalModel] interface.
	</p>		</p>
	<hr />		<hr />

Grau: /* Using existing models */

2012-02-02T15:00:58Z

Using existing models

← Older revision		Revision as of 15:00, 2 February 2012
Line 47:		Line 47:
	<p>		<p>
	'''Q: I have to decide, which model is best for my classification task. How do I assess different model combinations or classifiers?'''<br />		'''Q: I have to decide, which model is best for my classification task. How do I assess different model combinations or classifiers?'''<br />
	'''A:''' You can use the subclasses of [http://www.jstacs.de/api/de/jstacs/classifiers/assessment/ClassifierAssessment.html ClassifierAssessment], e.g. [http://www.jstacs.de/api/de/jstacs/classifiers/assessment/KFoldCrossValidation.html KFoldCrossValidation]. All [http://www.jstacs.de/api/de/jstacs/classifiers/assessment/ClassifierAssessment.html ClassifierAssessments] have a constructor that accepts an array of classifiers (or [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html TrainableStatisticalModels]). You can then use the [http://www.jstacs.de/api/de/jstacs/classifiers/assessment/ClassifierAssessment.html#assess(de.jstacs.~~classifier~~.~~MeasureParameters~~,%20de.jstacs.~~classifier~~.assessment.ClassifierAssessmentAssessParameterSet,%20de.jstacs.data.~~Sample~~...) assess] method to assess these classifiers on the same data using a number of pre-defined [http://www.jstacs.de/api/de/jstacs/~~classifier~~/~~MeasureParameters.Measure~~.html performance measures].~~<br />~~		'''A:''' You can use the subclasses of [http://www.jstacs.de/api/de/jstacs/classifiers/assessment/ClassifierAssessment.html ClassifierAssessment], e.g. [http://www.jstacs.de/api/de/jstacs/classifiers/assessment/KFoldCrossValidation.html KFoldCrossValidation]. All [http://www.jstacs.de/api/de/jstacs/classifiers/assessment/ClassifierAssessment.html ClassifierAssessments] have a constructor that accepts an array of classifiers (or [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html TrainableStatisticalModels]). You can then use the [http://www.jstacs.de/api/de/jstacs/classifiers/assessment/ClassifierAssessment.html#assess(de.jstacs.classifiers.performanceMeasures.NumericalPerformanceMeasureParameterSet,%20de.jstacs.classifiers.assessment.ClassifierAssessmentAssessParameterSet,%20de.jstacs.data.DataSet...) assess] method to assess these classifiers on the same data using a number of pre-defined [http://www.jstacs.de/api/de/jstacs/classifiers/performanceMeasures/NumericalPerformanceMeasureParameterSet.html performance measures].
	~~Also see the [[Performing a 10-fold cross validation \| code example]]~~
	</p>		</p>
	<hr />		<hr />
	<p>		<p>
	'''Q: How can store and load my model, classifier, ...?'''<br />		'''Q: How can store and load my model, classifier, ...?'''<br />
	'''A:''' All classes that implement [http://www.jstacs.de/api/de/jstacs/Storable.html Storable] have a method [http://www.jstacs.de/api/de/jstacs/Storable.html#toXML() toXML()] that returns a StringBuffer containing the instance as [http://en.wikipedia.org/wiki/XML XML]. Such classes should also have a proper constructor with a single argument StringBuffer. This can be used to create a new instance form a StringBuffer that contains an instance as XML. In addition, the class [http://www.jstacs.de/api/de/jstacs/io/FileManager.html FileManager] allows to read and write StringBuffers to the hard drive~~.<br />~~		'''A:''' All classes that implement [http://www.jstacs.de/api/de/jstacs/Storable.html Storable] have a method [http://www.jstacs.de/api/de/jstacs/Storable.html#toXML() toXML()] that returns a StringBuffer containing the instance as [http://en.wikipedia.org/wiki/XML XML]. Such classes should also have a proper constructor with a single argument StringBuffer. This can be used to create a new instance form a StringBuffer that contains an instance as XML. In addition, the class [http://www.jstacs.de/api/de/jstacs/io/FileManager.html FileManager] allows to read and write StringBuffers to the hard drive.
	~~Also see the [[Saving and loading a model \| code example]]~~.
	</p>		</p>
	<hr />		<hr />
Line 63:		Line 61:
	<hr />		<hr />
	<p>		<p>
	'''Q: I use Gibbs sampling in a class extending [http://www.jstacs.de/api/de/jstacs/~~models~~/mixture/~~AbstractMixtureModel~~.html ~~AbstractMixtureModel~~].'''<br />		'''Q: I use Gibbs sampling in a class extending [http://www.jstacs.de/api-2.0/de/jstacs/sequenceScores/statisticalModels/trainable/mixture/AbstractMixtureTrainSM.html AbstractMixtureTrainSM].'''<br />
	* '''Q<sub>1</sub>: Why does the sampling create files in temporary directory of Java?''' <br />		* '''Q<sub>1</sub>: Why does the sampling create files in temporary directory of Java?''' <br />
	* '''Q<sub>2</sub>: Will these files be deleted automatically, if they will not be used any more?'''<br />		* '''Q<sub>2</sub>: Will these files be deleted automatically, if they will not be used any more?'''<br />
	'''A:'''<br />		'''A:'''<br />
	* '''A<sub>1</sub>:''' These files are created for saving the sampled parameter temporarily. Java temp is used to minimize network load if you work on a cluster.<br />		* '''A<sub>1</sub>:''' These files are created for saving the sampled parameter temporarily. Java temp is used to minimize network load if you work on a cluster.<br />
	* '''A<sub>2</sub>:''' These files will be deleted if no reference to the mixture instance exists and the Garbage collector is called. Therefore it is recommended to [http://~~java~~.~~sun~~.com/~~j2se~~/~~1.5.0~~/docs/api/java/lang/System.html#gc() call the Garbage collector explicitly] at the end of any application.</p>		* '''A<sub>2</sub>:''' These files will be deleted if no reference to the mixture instance exists and the Garbage collector is called. Therefore it is recommended to [http://docs.oracle.com/javase/6/docs/api/java/lang/System.html#gc() call the Garbage collector explicitly] at the end of any application.</p>
	<hr />		<hr />

Grau: /* Using existing models */

2012-02-02T14:56:24Z

Using existing models

← Older revision		Revision as of 14:56, 2 February 2012
Line 42:		Line 42:
	<p>		<p>
	'''Q: I decided for two [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/differentiable/DifferentiableStatisticalModel.html DifferentiableStatisticalModels]. How do I learn them and classify new data?'''<br />		'''Q: I decided for two [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/differentiable/DifferentiableStatisticalModel.html DifferentiableStatisticalModels]. How do I learn them and classify new data?'''<br />
	'''A:''' You can create a new [http://www.jstacs.de/api/de/jstacs/classifiers/differentiableSequenceScoreBased/gendismix/GenDisMixClassifier.html GenDisMixClassifier] from your models and use its [http://www.jstacs.de/api/de/jstacs/classifier/scoringFunctionBased/ScoreClassifier.html#train(de.jstacs.data.Sample%5B%5D,%20double%5B%5D%5B%5D) train] and [http://www.jstacs.de/api/de/jstacs/classifiers/differentiableSequenceScoreBased/gendismix/GenDisMixClassifier.html#classify(de.jstacs.data.Sequence)) classify] methods.		'''A:''' You can create a new [http://www.jstacs.de/api/de/jstacs/classifiers/differentiableSequenceScoreBased/gendismix/GenDisMixClassifier.html GenDisMixClassifier] from your models and use its [http://www.jstacs.de/api/de/jstacs/classifier/scoringFunctionBased/ScoreClassifier.html#train(de.jstacs.data.Sample%5B%5D,%20double%5B%5D%5B%5D) train] and [http://www.jstacs.de/api/de/jstacs/classifiers/differentiableSequenceScoreBased/gendismix/GenDisMixClassifier.html#classify(de.jstacs.data.sequences.Sequence)) classify] methods.
	</p>		</p>
	<hr />		<hr />
	<p>		<p>
	'''Q: I have to decide, which model is best for my classification task. How do I assess different model combinations or classifiers?'''<br />		'''Q: I have to decide, which model is best for my classification task. How do I assess different model combinations or classifiers?'''<br />
	'''A:''' You can use the subclasses of [http://www.jstacs.de/api/de/jstacs/~~classifier~~/assessment/ClassifierAssessment.html ClassifierAssessment], e.g. [http://www.jstacs.de/api/de/jstacs/~~classifier~~/assessment/KFoldCrossValidation.html KFoldCrossValidation]. All [http://www.jstacs.de/api/de/jstacs/~~classifier~~/assessment/ClassifierAssessment.html ClassifierAssessments] have a constructor that accepts an array of classifiers (or [http://www.jstacs.de/api/de/jstacs/~~models~~/~~Model~~.html ~~Models~~]). You can then use the [http://www.jstacs.de/api/de/jstacs/~~classifier~~/assessment/ClassifierAssessment.html#assess(de.jstacs.classifier.MeasureParameters,%20de.jstacs.classifier.assessment.ClassifierAssessmentAssessParameterSet,%20de.jstacs.data.Sample...) assess] method to assess these classifiers on the same data using a number of pre-defined [http://www.jstacs.de/api/de/jstacs/classifier/MeasureParameters.Measure.html performance measures].<br />		'''A:''' You can use the subclasses of [http://www.jstacs.de/api/de/jstacs/classifiers/assessment/ClassifierAssessment.html ClassifierAssessment], e.g. [http://www.jstacs.de/api/de/jstacs/classifiers/assessment/KFoldCrossValidation.html KFoldCrossValidation]. All [http://www.jstacs.de/api/de/jstacs/classifiers/assessment/ClassifierAssessment.html ClassifierAssessments] have a constructor that accepts an array of classifiers (or [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html TrainableStatisticalModels]). You can then use the [http://www.jstacs.de/api/de/jstacs/classifiers/assessment/ClassifierAssessment.html#assess(de.jstacs.classifier.MeasureParameters,%20de.jstacs.classifier.assessment.ClassifierAssessmentAssessParameterSet,%20de.jstacs.data.Sample...) assess] method to assess these classifiers on the same data using a number of pre-defined [http://www.jstacs.de/api/de/jstacs/classifier/MeasureParameters.Measure.html performance measures].<br />
	Also see the [[Performing a 10-fold cross validation \| code example]]		Also see the [[Performing a 10-fold cross validation \| code example]]
	</p>		</p>

Grau: /* Using existing models */

2012-02-02T14:55:00Z

Using existing models

← Older revision		Revision as of 14:55, 2 February 2012
Line 41:		Line 41:
	<hr />		<hr />
	<p>		<p>
	'''Q: I decided for two [http://www.jstacs.de/api/de/jstacs/~~scoringFunctions~~/~~ScoringFunction~~.html ~~ScoringFunctions~~]. How do I learn them and classify new data?'''<br />		'''Q: I decided for two [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/differentiable/DifferentiableStatisticalModel.html DifferentiableStatisticalModels]. How do I learn them and classify new data?'''<br />
	'''A:''' You can create a new [http://www.jstacs.de/api/de/jstacs/~~classifier~~/~~scoringFunctionBased~~/~~ScoreClassifier~~.html ~~ScoreClassifier~~] from your ~~scoring functions~~ and use its [http://www.jstacs.de/api/de/jstacs/classifier/scoringFunctionBased/ScoreClassifier.html#train(de.jstacs.data.Sample%5B%5D,%20double%5B%5D%5B%5D) train] and [http://www.jstacs.de/api/de/jstacs/~~classifier~~/~~AbstractScoreBasedClassifier~~.html#classify(de.jstacs.data.Sequence)) classify] methods.		'''A:''' You can create a new [http://www.jstacs.de/api/de/jstacs/classifiers/differentiableSequenceScoreBased/gendismix/GenDisMixClassifier.html GenDisMixClassifier] from your models and use its [http://www.jstacs.de/api/de/jstacs/classifier/scoringFunctionBased/ScoreClassifier.html#train(de.jstacs.data.Sample%5B%5D,%20double%5B%5D%5B%5D) train] and [http://www.jstacs.de/api/de/jstacs/classifiers/differentiableSequenceScoreBased/gendismix/GenDisMixClassifier.html#classify(de.jstacs.data.Sequence)) classify] methods.
	</p>		</p>
	<hr />		<hr />

Grau: /* Using existing models */

2012-02-02T14:52:48Z

Using existing models

← Older revision		Revision as of 14:52, 2 February 2012
Line 37:		Line 37:
	<p>		<p>
	'''Q: I decided for two [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html TrainableStatisticalModels]. How do I learn them and classify new data?'''<br />		'''Q: I decided for two [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html TrainableStatisticalModels]. How do I learn them and classify new data?'''<br />
	'''A:''' You can create a new [http://www.jstacs.de/api/de/jstacs/classifiers/trainSMBased/TrainSMBasedClassifier.html TrainSMBasedClassifier] from your models and use its [http://www.jstacs.de/api/de/jstacs/~~classifier~~/~~modelBased~~/~~ModelBasedClassifier~~.html#train(de.jstacs.data.~~Sample~~%5B%5D,%20double%5B%5D%5B%5D) train] and [http://www.jstacs.de/api/de/jstacs/~~classifier~~/~~modelBased~~/~~ModelBasedClassifier~~.html#classify(de.jstacs.data.~~Sample~~) classify] methods. If you only want to learn a model from data, e.g. to sample new sequences, you can also directly use the [http://www.jstacs.de/api/de/jstacs/~~models~~/~~Model~~.html#train(de.jstacs.data.~~Sample~~) train] method of the [http://www.jstacs.de/api/de/jstacs/~~models~~/~~Model~~.html ~~Model].<br />~~		'''A:''' You can create a new [http://www.jstacs.de/api/de/jstacs/classifiers/trainSMBased/TrainSMBasedClassifier.html TrainSMBasedClassifier] from your models and use its [http://www.jstacs.de/api/de/jstacs/classifiers/trainSMBased/TrainSMBasedClassifier.html#train(de.jstacs.data.DataSet%5B%5D,%20double%5B%5D%5B%5D) train] and [http://www.jstacs.de/api/de/jstacs/classifiers/trainSMBased/TrainSMBasedClassifier.html#classify(de.jstacs.data.DataSet) classify] methods. If you only want to learn a model from data, e.g. to sample new sequences, you can also directly use the [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html#train(de.jstacs.data.DataSet) train] method of the [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html TrainableStatisticalModel].
	~~Also see the [[Training a classifier and classifying new sequences \| code example]~~].
	</p>		</p>
	<hr />		<hr />

Grau: /* Using existing models */

2012-02-02T14:51:05Z

Using existing models

← Older revision		Revision as of 14:51, 2 February 2012
Line 37:		Line 37:
	<p>		<p>
	'''Q: I decided for two [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html TrainableStatisticalModels]. How do I learn them and classify new data?'''<br />		'''Q: I decided for two [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html TrainableStatisticalModels]. How do I learn them and classify new data?'''<br />
	'''A:''' You can create a new [http://www.jstacs.de/api/de/jstacs/~~classifier~~/trainSMBased/TrainSMBasedClassifier.html TrainSMBasedClassifier] from your models and use its [http://www.jstacs.de/api/de/jstacs/classifier/modelBased/ModelBasedClassifier.html#train(de.jstacs.data.Sample%5B%5D,%20double%5B%5D%5B%5D) train] and [http://www.jstacs.de/api/de/jstacs/classifier/modelBased/ModelBasedClassifier.html#classify(de.jstacs.data.Sample) classify] methods. If you only want to learn a model from data, e.g. to sample new sequences, you can also directly use the [http://www.jstacs.de/api/de/jstacs/models/Model.html#train(de.jstacs.data.Sample) train] method of the [http://www.jstacs.de/api/de/jstacs/models/Model.html Model].<br />		'''A:''' You can create a new [http://www.jstacs.de/api/de/jstacs/classifiers/trainSMBased/TrainSMBasedClassifier.html TrainSMBasedClassifier] from your models and use its [http://www.jstacs.de/api/de/jstacs/classifier/modelBased/ModelBasedClassifier.html#train(de.jstacs.data.Sample%5B%5D,%20double%5B%5D%5B%5D) train] and [http://www.jstacs.de/api/de/jstacs/classifier/modelBased/ModelBasedClassifier.html#classify(de.jstacs.data.Sample) classify] methods. If you only want to learn a model from data, e.g. to sample new sequences, you can also directly use the [http://www.jstacs.de/api/de/jstacs/models/Model.html#train(de.jstacs.data.Sample) train] method of the [http://www.jstacs.de/api/de/jstacs/models/Model.html Model].<br />
	Also see the [[Training a classifier and classifying new sequences \| code example]].		Also see the [[Training a classifier and classifying new sequences \| code example]].
	</p>		</p>

Grau: /* Using existing models */

2012-02-02T14:50:37Z

Using existing models

← Older revision		Revision as of 14:50, 2 February 2012
Line 28:		Line 28:

	== Using existing models ==		== Using existing models ==
	Also have a look at ~~code examples~~ [[~~Training a classifier and classifying new sequences]], [[Performing a 10-fold cross validation]] and [[Saving and loading a model~~]].		Also have a look at the [[Code examples]].
	<p>		<p>
	'''Q: Where do I find a list of the models currently implemented in Jstacs?'''<br />		'''Q: Where do I find a list of the models currently implemented in Jstacs?'''<br />
	'''A:''' All generative models in Jstacs implement the [http://www.jstacs.de/api/de/jstacs/~~models~~/~~Model~~.html ~~Model~~] interface.<br />		'''A:''' All generative models in Jstacs implement the [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html TrainableStatisticalModel] interface.<br />
	All discriminative models in Jstacs implement the [http://www.jstacs.de/api/de/jstacs/~~scoringFunctions~~/~~ScoringFunction~~.html ~~ScoringFunction~~] interface. You find all the existing implementations in the list of implementing classes of these two interfaces.		All discriminative models in Jstacs implement the [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/differentiable/DifferentiableStatisticalModel.html DifferentiableStatisticalModel] interface. You find all the existing implementations in the list of implementing classes of these two interfaces.
	</p>		</p>
	<hr />		<hr />
	<p>		<p>
	'''Q: I decided for two [http://www.jstacs.de/api/de/jstacs/~~models~~/~~Model~~.html ~~Models~~]. How do I learn them and classify new data?'''<br />		'''Q: I decided for two [http://www.jstacs.de/api/de/jstacs/sequenceScores/statisticalModels/trainable/TrainableStatisticalModel.html TrainableStatisticalModels]. How do I learn them and classify new data?'''<br />
	'''A:''' You can create a new [http://www.jstacs.de/api/de/jstacs/classifier/~~modelBased~~/~~ModelBasedClassifier~~.html ~~ModelBasedClassifier~~] from your models and use its [http://www.jstacs.de/api/de/jstacs/classifier/modelBased/ModelBasedClassifier.html#train(de.jstacs.data.Sample%5B%5D,%20double%5B%5D%5B%5D) train] and [http://www.jstacs.de/api/de/jstacs/classifier/modelBased/ModelBasedClassifier.html#classify(de.jstacs.data.Sample) classify] methods. If you only want to learn a model from data, e.g. to sample new sequences, you can also directly use the [http://www.jstacs.de/api/de/jstacs/models/Model.html#train(de.jstacs.data.Sample) train] method of the [http://www.jstacs.de/api/de/jstacs/models/Model.html Model].<br />		'''A:''' You can create a new [http://www.jstacs.de/api/de/jstacs/classifier/trainSMBased/TrainSMBasedClassifier.html TrainSMBasedClassifier] from your models and use its [http://www.jstacs.de/api/de/jstacs/classifier/modelBased/ModelBasedClassifier.html#train(de.jstacs.data.Sample%5B%5D,%20double%5B%5D%5B%5D) train] and [http://www.jstacs.de/api/de/jstacs/classifier/modelBased/ModelBasedClassifier.html#classify(de.jstacs.data.Sample) classify] methods. If you only want to learn a model from data, e.g. to sample new sequences, you can also directly use the [http://www.jstacs.de/api/de/jstacs/models/Model.html#train(de.jstacs.data.Sample) train] method of the [http://www.jstacs.de/api/de/jstacs/models/Model.html Model].<br />
	Also see the [[Training a classifier and classifying new sequences \| code example]].		Also see the [[Training a classifier and classifying new sequences \| code example]].
	</p>		</p>

Grau at 14:48, 2 February 2012

2012-02-02T14:48:00Z

← Older revision		Revision as of 14:48, 2 February 2012
Line 18:		Line 18:
	<p>		<p>
	'''Q: How can I load my own data?'''<br />		'''Q: How can I load my own data?'''<br />
	'''A:''' If your sequences are stored either in plain text or in FastA format, you can directly [http://www.jstacs.de/api/de/jstacs/data/DataSet.html#DataSet(de.jstacs.data.AlphabetContainer,%20de.jstacs.io.~~StringExtractor~~,%20java.lang.String) create] a new [http://www.jstacs.de/api/de/jstacs/data/DataSet.html DataSet] from the file.		'''A:''' If your sequences are stored either in plain text or in FastA format, you can directly [http://www.jstacs.de/api/de/jstacs/data/DataSet.html#DataSet(de.jstacs.data.AlphabetContainer,%20de.jstacs.io.AbstractStringExtractor,%20java.lang.String) create] a new [http://www.jstacs.de/api/de/jstacs/data/DataSet.html DataSet] from the file.
	</p>		</p>
	<hr />		<hr />

Grau: /* Handling data */

2012-02-02T14:46:51Z

Handling data

← Older revision		Revision as of 14:46, 2 February 2012
Line 13:		Line 13:
	<p>		<p>
	'''Q: How can I create a simple sequence?'''<br />		'''Q: How can I create a simple sequence?'''<br />
	'''A:''' Try to use the [http://www.jstacs.de/api/de/jstacs/data/~~sequencesSequence~~.html#create(de.jstacs.data.AlphabetContainer,%20java.lang.String) create method] of [http://www.jstacs.de/api/de/jstacs/data/sequences/Sequence.html Sequence], e.q. Sequence.create( new AlphabetContainer( new DNAAlphabet() ), "ACGTACGT" );		'''A:''' Try to use the [http://www.jstacs.de/api/de/jstacs/data/sequences/Sequence.html#create(de.jstacs.data.AlphabetContainer,%20java.lang.String) create method] of [http://www.jstacs.de/api/de/jstacs/data/sequences/Sequence.html Sequence], e.q. Sequence.create( new AlphabetContainer( new DNAAlphabet() ), "ACGTACGT" );
	</p>		</p>
	<hr />		<hr />

Grau at 14:46, 2 February 2012

2012-02-02T14:46:38Z

← Older revision		Revision as of 14:46, 2 February 2012
Line 13:		Line 13:
	<p>		<p>
	'''Q: How can I create a simple sequence?'''<br />		'''Q: How can I create a simple sequence?'''<br />
	'''A:''' Try to use the [http://www.jstacs.de/api/de/jstacs/data/~~Sequence~~.html#create(de.jstacs.data.AlphabetContainer,%20java.lang.String) create method] of [http://www.jstacs.de/api/de/jstacs/data/Sequence.html Sequence], e.q. Sequence.create( new AlphabetContainer( new DNAAlphabet() ), "ACGTACGT" );		'''A:''' Try to use the [http://www.jstacs.de/api/de/jstacs/data/sequencesSequence.html#create(de.jstacs.data.AlphabetContainer,%20java.lang.String) create method] of [http://www.jstacs.de/api/de/jstacs/data/sequences/Sequence.html Sequence], e.q. Sequence.create( new AlphabetContainer( new DNAAlphabet() ), "ACGTACGT" );
	</p>		</p>
	<hr />		<hr />
	<p>		<p>
	'''Q: How can I load my own data?'''<br />		'''Q: How can I load my own data?'''<br />
	'''A:''' If your sequences are stored either in plain text or in FastA format, you can directly [http://www.jstacs.de/api/de/jstacs/data/~~Sample~~.html#~~Sample~~(de.jstacs.data.AlphabetContainer,%20de.jstacs.io.StringExtractor,%20java.lang.String) create] a new [http://www.jstacs.de/api/de/jstacs/data/~~Sample~~.html ~~Sample~~] from the file.		'''A:''' If your sequences are stored either in plain text or in FastA format, you can directly [http://www.jstacs.de/api/de/jstacs/data/DataSet.html#DataSet(de.jstacs.data.AlphabetContainer,%20de.jstacs.io.StringExtractor,%20java.lang.String) create] a new [http://www.jstacs.de/api/de/jstacs/data/DataSet.html DataSet] from the file.
	</p>		</p>
	<hr />		<hr />