def statistics_kmm (n,d): from modshogun import RealFeatures from modshogun import DataGenerator from modshogun import GaussianKernel, MSG_DEBUG from modshogun import KernelMeanMatching from modshogun import Math # init seed for reproducability Math.init_random(1) random.seed(1); data = random.randn(d,n) # create shogun feature representation features=RealFeatures(data) # use a kernel width of sigma=2, which is 8 in SHOGUN's parametrization # which is k(x,y)=exp(-||x-y||^2 / tau), in constrast to the standard # k(x,y)=exp(-||x-y||^2 / (2*sigma^2)), so tau=2*sigma^2 kernel=GaussianKernel(10,8) kernel.init(features,features) kmm = KernelMeanMatching(kernel,array([0,1,2,3,7,8,9],dtype=int32),array([4,5,6],dtype=int32)) w = kmm.compute_weights() #print w return w
def statistics_kmm(n, d):
from modshogun import RealFeatures
from modshogun import DataGenerator
from modshogun import GaussianKernel, MSG_DEBUG
from modshogun import KernelMeanMatching
from modshogun import Math
# init seed for reproducability
Math.init_random(1)
random.seed(1)
data = random.randn(d, n)
# create shogun feature representation
features = RealFeatures(data)
# use a kernel width of sigma=2, which is 8 in SHOGUN's parametrization
# which is k(x,y)=exp(-||x-y||^2 / tau), in constrast to the standard
# k(x,y)=exp(-||x-y||^2 / (2*sigma^2)), so tau=2*sigma^2
kernel = GaussianKernel(10, 8)
kernel.init(features, features)
kmm = KernelMeanMatching(kernel, array([0, 1, 2, 3, 7, 8, 9], dtype=int32),
array([4, 5, 6], dtype=int32))
w = kmm.compute_weights()
#print w
return w
def train_kmm(self):
width = float(self.sigma.text())
degree = int(self.degree.text())
self.axes.clear()
self.axes.grid(True)
self.axes.plot(self.data.x1_test, self.data.x2_test, 'ro')
self.axes.plot(self.data.x1_train, self.data.x2_train, 'bo')
# train kmm
labels = self.data.get_labels()
lab = BinaryLabels(labels)
features = self.data.get_examples()
train = RealFeatures(features)
nTrain=len(self.data.x1_train);
nTest=len(self.data.x1_test);
trainI=numpy.array(range(nTrain), dtype=numpy.int32)
testI=numpy.array(range(nTrain,nTest+nTrain),dtype=numpy.int32)
kernel_name = self.kernel_combo.currentText()
print "current kernel is %s" % (kernel_name)
if kernel_name == "LinearKernel":
gk = LinearKernel(train, train)
gk.set_normalizer(IdentityKernelNormalizer())
elif kernel_name == "PolynomialKernel":
gk = PolyKernel(train, train, degree, True)
gk.set_normalizer(IdentityKernelNormalizer())
elif kernel_name == "GaussianKernel":
gk = GaussianKernel(train, train, width)
kmm = KernelMeanMatching(gk, trainI, testI)
w = kmm.compute_weights()
print 'Weights'
print w
self.axes.clear()
self.axes.grid(True)
self.axes.plot(self.data.x1_test, self.data.x2_test, 'ro')
m_size=numpy.array(w*1000, dtype=numpy.int32)
self.axes.scatter(self.data.x1_train, self.data.x2_train, s=m_size)
self.axes.set_xlim((-5,5))
self.axes.set_ylim((-5,5))
self.canvas.draw()
def train_kmm(self):
width = float(self.sigma.text())
degree = int(self.degree.text())
self.axes.clear()
self.axes.grid(True)
self.axes.plot(self.data.x1_test, self.data.x2_test, 'ro')
self.axes.plot(self.data.x1_train, self.data.x2_train, 'bo')
# train kmm
labels = self.data.get_labels()
lab = BinaryLabels(labels)
features = self.data.get_examples()
train = RealFeatures(features)
nTrain = len(self.data.x1_train)
nTest = len(self.data.x1_test)
trainI = numpy.array(range(nTrain), dtype=numpy.int32)
testI = numpy.array(range(nTrain, nTest + nTrain), dtype=numpy.int32)
kernel_name = self.kernel_combo.currentText()
print "current kernel is %s" % (kernel_name)
if kernel_name == "LinearKernel":
gk = LinearKernel(train, train)
gk.set_normalizer(IdentityKernelNormalizer())
elif kernel_name == "PolynomialKernel":
gk = PolyKernel(train, train, degree, True)
gk.set_normalizer(IdentityKernelNormalizer())
elif kernel_name == "GaussianKernel":
gk = GaussianKernel(train, train, width)
kmm = KernelMeanMatching(gk, trainI, testI)
w = kmm.compute_weights()
print 'Weights'
print w
self.axes.clear()
self.axes.grid(True)
self.axes.plot(self.data.x1_test, self.data.x2_test, 'ro')
m_size = numpy.array(w * 1000, dtype=numpy.int32)
self.axes.scatter(self.data.x1_train, self.data.x2_train, s=m_size)
self.axes.set_xlim((-5, 5))
self.axes.set_ylim((-5, 5))
self.canvas.draw()
