def decision_function(self, X):
if self.is_fitted == False:
raise NotFittedError(
"This EasyMKL instance is not fitted yet. Call 'fit' with appropriate arguments before using this method."
)
if self.kernel == 'precomputed':
K = check_KL_Y(X, self.Y)
else:
X = check_array(X,
accept_sparse='csr',
dtype=np.float64,
order="C")
if X.shape[1] != self.n_f:
raise ValueError("The number of feature in X not correspond")
#K = self.K.set_test(X)
K = kernel_list(self.X, X, self.K)
if self.multiclass_ == True:
return self.cls.decision_function(X)
YY = matrix(np.diag(list(matrix(self.Y))))
ker_matrix = matrix(summation(K, self.weights))
z = ker_matrix * YY * self.gamma
z = z - self.bias
return np.array(list(z))
def _check_test(self,X):
if self.is_fitted == False:
raise NotFittedError("This EasyMKL instance is not fitted yet. Call 'fit' with appropriate arguments before using this method.")
if self.kernel == 'precomputed':
KL = check_KL_Y(X,self.Y)
else:
X = check_array(X, accept_sparse='csr', dtype=np.float64, order="C")
if X.shape[1] != self.X.shape[1]:#self.n_f:
raise ValueError("The number of feature in X not correspond")
KL = self.gen(self.X,X).make_a_list(self.n_kernels).to_array()
return KL
def _prepare(self,X,Y):
'''preprocess data before training'''
check_classification_targets(Y)
self.classes_ = np.unique(Y)
if len(self.classes_) < 2:
raise ValueError("The number of classes has to be almost 2; got ", len(self.classes_))
self.multiclass_ = len(self.classes_) > 2
KL = process_list(X,self.generator) # X can be a samples matrix or Kernels List
self.KL, self.Y = check_KL_Y(KL,Y)
self.n_kernels = len(self.KL)
return
def _prepare(self,X,Y):
'''preprocess data before training'''
check_classification_targets(Y)
self.classes_ = np.unique(Y)
if len(self.classes_) < 2:
raise ValueError("The number of classes has to be almost 2; got ", len(self.classes_))
self.multiclass_ = len(self.classes_) > 2
KL = process_list(X,self.generator) # X can be a samples matrix or Kernels List
self.KL, self.Y = check_KL_Y(KL,Y)
self.n_kernels = len(self.KL)
return
def _input_checks(self, X, Y):
'''valida l'input dell'algoritmo e fa un po di preprocessing'''
if self.kernel == 'precomputed':
self.KL = check_KL_Y(X,Y)
else:
X,Y = validation.check_X_y(X, Y, dtype=np.float64, order='C', accept_sparse='csr')
self.KL = self.gen(X).make_a_list(self.n_kernels)
check_classification_targets(Y)
self.Y = Y
self.X = X
self.classes_ = np.unique(Y)
if len(self.classes_) < 2:
raise ValueError("The number of classes has to be almost 2; got ", len(self.classes_))
if self.tracenorm:
for k in self.KL:
self.traces.append(traceN(k))
def _input_checks(self, X, Y):
'''valida l'input dell'algoritmo e fa un po di preprocessing'''
if self.kernel == 'precomputed':
self.K = check_KL_Y(X, Y)
elif self.gen != None:
X, Y = validation.check_X_y(X,
Y,
dtype=np.float64,
order='C',
accept_sparse='csr')
self.K = self.gen.make_a_list(self.n_kernels)
else:
random.seed(1)
X, Y = validation.check_X_y(X,
Y,
dtype=np.float64,
order='C',
accept_sparse='csr')
self.K = random_generator(X).make_a_list(
self.n_kernels) #.to_array() #da sistemare i parametri
#self.K = poly_generator(X).make_a_list(20)#.to_array()
check_classification_targets(Y)
self.Y = Y
self.X = X
self.classes_ = np.unique(Y)
if len(self.classes_) < 2:
raise ValueError("The number of classes has to be almost 2; got ",
len(self.classes_))
self.traces = []
#for k in self.K:
# self.traces.append(traceN(k))
if self.tracenorm and False:
for k in self.K:
self.traces.append(traceN(k))
from MKL.utils import kernel_list
if not type(self.K).__name__ == 'ndarray': #== kernel_list:
#print 'k_list'
self.K /= self.traces
else:
#print 'array'
c = [[[i for j in range(len(Y))] for k in range(len(Y))]
for i in self.traces]
self.K /= np.array(c)
del c
self.n_f = X.shape[1]