def __init__(self, alpha, threshold, sample, **kwargs):
r"""See ``SVMClassifier`` for full documentation.
"""
Classifier.__init__(self)
num_patterns = len(sample)
check_svm_classification_sample(sample)
self.dim = len(sample[0].pattern)
if len(alpha) != num_patterns:
raise ValueError('The supplied sample and multipliers vector do \
not have the same size')
self.sv_indices = [i for i in range(len(alpha)) if alpha[i] != 0]
self.support_vectors = [sample[i].pattern for i in self.sv_indices]
self.signed_alphas = [alpha[i] * sample[i].label
for i in self.sv_indices]
self.threshold = threshold
try:
self.kernel = kwargs['kernel']
except KeyError:
self.kernel = Kernel.get_default()
def __repr__(self):
alpha = [abs(a) for a in self.signed_alphas]
# was
# map(abs, self.signed_alphas)
patterns = self.support_vectors
labels = [sign(a) for a in self.signed_alphas]
# was
# labels = map(sign, self.signed_alphas)
sample = [LabeledExample(*pl) for pl in zip(patterns, labels)]
# was
# sample = map(lambda x: LabeledExample(*x), zip(patterns, labels))
result = 'SVMClassifier(' + str(alpha) + ', '
result += str(self.threshold) + ', '
result += str(sample.__repr__())
if self.kernel != Kernel.get_default():
result += ', kernel = ' + str(self.kernel.__repr__())
result += ')'
return result
