Home       Trees       Indices       Help   
PyMVPA: Python MultiVariate Pattern Analysis
Package mvpa :: Package clfs :: Module transerror :: Class ClassifierError
[hide private]
[frames] | no frames]

Class ClassifierError

source code

Compute (or return) some error of a (trained) classifier on a dataset.
Nested Classes [hide private]

Inherited from misc.state.ClassWithCollections: __metaclass__

Instance Methods [hide private]
 
__init__(self, clf, labels=None, train=True, **kwargs)
Initialization.		 source code
 
__copy__(self)
TODO: think... may be we need to copy self.clf		 source code
 
_precall(self, testdataset, trainingdataset=None)
Generic part which trains the classifier if necessary		 source code
 
_call(self, testdataset, trainingdataset=None) source code
 
_postcall(self, testdataset, trainingdataset=None, error=None) source code
 
__call__(self, testdataset, trainingdataset=None)
Compute the transfer error for a certain test dataset.		 source code
 
untrain(self)
Untrain the *Error which relies on the classifier		 source code
 
clf(self) source code
 
labels(self) source code

Inherited from misc.state.ClassWithCollections: __getattribute__, __new__, __repr__, __setattr__, __str__, reset

Inherited from object: __delattr__, __format__, __hash__, __reduce__, __reduce_ex__, __sizeof__, __subclasshook__

Class Variables [hide private]
  confusion = StateVariable(enabled= False)
TODO Think that labels might be also symbolic thus can't directly be indicies of the array
  training_confusion = StateVariable(enabled= False, doc= "Proxy...
  __doc__ = enhancedDocString('ClassifierError', locals(), Class...

Inherited from misc.state.ClassWithCollections: _DEV__doc__, descr

Instance Variables [hide private]
  _labels
Labels to add on top to existing in testing data
  __train
Either to train classifier if trainingdata is provided
Properties [hide private]

Inherited from object: __class__

Method Details [hide private]

__init__(self, clf, labels=None, train=True, **kwargs)
(Constructor)

source code 
Initialization.
Parameters:
  • clf (Classifier) - Either trained or untrained classifier
  • labels (list) - if provided, should be a set of labels to add on top of the ones present in testdata
  • train (bool) - unless train=False, classifier gets trained if trainingdata provided to __call__
Overrides: object.__init__

__call__(self, testdataset, trainingdataset=None)
(Call operator)

source code 

Compute the transfer error for a certain test dataset.

If trainingdataset is not None the classifier is trained using the provided dataset before computing the transfer error. Otherwise the classifier is used in it's current state to make the predictions on the test dataset.

Returns a scalar value of the transfer error.

clf(self)

source code 
Decorators:
  • @property

labels(self)

source code 
Decorators:
  • @property

Class Variable Details [hide private]

training_confusion

Value:
StateVariable(enabled= False, doc= "Proxy training_confusion from unde\
rlying classifier.")

__doc__

Value:
enhancedDocString('ClassifierError', locals(), ClassWithCollections)

   Home       Trees       Indices       Help   
PyMVPA: Python MultiVariate Pattern Analysis
Generated by Epydoc 3.0.1 on Mon Apr 23 23:09:23 2012 http://epydoc.sourceforge.net