Method __init__

pattern/vector/__init__.py:2143–2188

Support Vector Machine (SVM) is a supervised learning method where training documents are represented as points in n-dimensional space. The SVM constructs a number of hyperplanes that subdivide the space. Optional parameters: - type = CLASSIFICA

(self, *args, **kwargs)

class SVM(Classifier):
    
    def __init__(self, *args, **kwargs):
        """ Support Vector Machine (SVM) is a supervised learning method 
            where training documents are represented as points in n-dimensional space.
            The SVM constructs a number of hyperplanes that subdivide the space.
            Optional parameters:
            -      type = CLASSIFICATION, 
            -    kernel = LINEAR, 
            -    degree = 3, 
            -     gamma = 1/len(SVM.features), 
            -    coeff0 = 0,
            -      cost = 1, 
            -   epsilon = 0.01, 
            -     cache = 100, 
            - shrinking = True
        """
        import svm
        self._svm = svm
        # Cached LIBSVM or LIBLINEAR model:
        self._model = None
        # SVM.extensions is a tuple of extension modules that can be used.
        # By default, LIBLINEAR will be used for linear SVC (it is faster).
        # If you do not want to use LIBLINEAR, use SVM(extension=LIBSVM).
        self._extensions = \
            kwargs.get("extensions", 
            kwargs.get("extension", (LIBSVM, LIBLINEAR)))
        # Optional parameters are read-only:
        if len(args) > 0: 
            kwargs.setdefault(  "type", args[0])
        if len(args) > 1: 
            kwargs.setdefault("kernel", args[1])
        if len(args) > 2: 
            kwargs.setdefault("degree", args[2])
        for k1, k2, v in (
            (       "type", "s", CLASSIFICATION),
            (     "kernel", "t", LINEAR),
            (     "degree", "d", 3),   # For POLYNOMIAL.
            (      "gamma", "g", 0),   # For POLYNOMIAL + RADIAL.
            (     "coeff0", "r", 0),   # For POLYNOMIAL.
            (       "cost", "c", 1),   # Can be optimized with gridsearch().
            (    "epsilon", "p", 0.1),
            (         "nu", "n", 0.5),
            (      "cache", "m", 100), # MB
            (  "shrinking", "h", True)):
                v = kwargs.get(k2, kwargs.get(k1, v))
                setattr(self, "_"+k1, v)
        Classifier.__init__(self, train=kwargs.get("train", []), baseline=FREQUENCY)
    
    @property
    def extension(self):