Home       Trees       Indices       Help   
Package pyvision :: Package vector :: Module SVM :: Class SVM
[hide private]
[frames] | no frames]

Class SVM

source code

VectorClassifier.VectorClassifier --+
                                    |
                                   SVM
Instance Methods [hide private]
 
__init__(self, svm_type=TYPE_SVC, kernel=KERNEL_RBF, svr_epsilon=0.1, nu=0.5, random_seed=None, validation_size=0.33, **kwargs)
Create an svm.		 source code
 
__getstate__(self)
This function is neccessary for pickling		 source code
 
__setstate__(self, state)
This function is neccessary for pickling		 source code
 
trainClassifer(self, labels, vectors, ilog=None, verbose=False, callback=None, C_range=2.0**np.arange(-5,16,1), G_range=2.0**np.arange(-15,4,1))
Do not call this function instead call train.		 source code
 
train_SVC_RBF(self, labels, vectors, verbose, C_range, G_range, callback=None)
Private use only		 source code
 
train_SVR_RBF(self, labels, vectors, verbose, C_range, G_range, callback=None)
Private use only		 source code
 
train_SVC_Linear(self, labels, vectors, verbose, C_range, callback=None)
Private use only.		 source code
 
train_SVR_Linear(self, labels, vectors, verbose, C_range, callback=None)
Private use only		 source code
 
predictValue(self, data, ilog=None)
Please call predict instead.		 source code
 
predictSVMProbability(self, data) source code
 
predictSVMValues(self, data) source code

Inherited from VectorClassifier.VectorClassifier: addTraining, invertClass, invertReg, normalizeVector, predict, train, trainNormalization

Method Details [hide private]

__init__(self, svm_type=TYPE_SVC, kernel=KERNEL_RBF, svr_epsilon=0.1, nu=0.5, random_seed=None, validation_size=0.33, **kwargs)
(Constructor)

source code 

Create an svm.

Make sure you choose "classifacition" or "regression". Other parameters control features of the SVM.

also passes keyword args to VectorClassifier

Overrides: VectorClassifier.VectorClassifier.__init__

trainClassifer(self, labels, vectors, ilog=None, verbose=False, callback=None, C_range=2.0**np.arange(-5,16,1), G_range=2.0**np.arange(-15,4,1))

source code 

Do not call this function instead call train.

Overrides: VectorClassifier.VectorClassifier.trainClassifer

predictValue(self, data, ilog=None)

source code 

Please call predict instead.

Overrides: VectorClassifier.VectorClassifier.predictValue

   Home       Trees       Indices       Help   
Generated by Epydoc 3.0.1 on Tue Apr 21 14:31:59 2015 http://epydoc.sourceforge.net