Package PyML :: Package classifiers :: Module svm
[frames] | no frames]

Module svm

source code

Classes
  SVM
An SVM classifier class.
  SVR
A class for SVM regression (libsvm wrapper).
  OneClassSVM
wrapper for the libsvm one-class SVM
  SVC
  SVModel
  LinearSVModel
Functions
 
loadSVM(fileName, **args)
returns a trained SVM object constructed from a saved SVM model.
source code
 
runMySMO(svmInstance, data) source code
 
runGist(classifier, data) source code
 
runGradientDescent(classifier, data) source code
Variables
  containersNotSupported = ['PySparseDataSet', 'PyVectorDataSet']
Function Details

loadSVM(fileName, **args)

source code 

returns a trained SVM object constructed from a saved SVM model.

The saved SVM model stores the support vectors in sparse vector format. When creating the model it then represents the support vectors in some dataset container. The type of the container needs to agree with the type of dataset of your test data. By default the support vectors are represented using the SparseDataSet container. You can set this using the 'datasetClass' keyword argument e.g. datasetClass = SparseDataSet