def train_svm(self):
width = float(self.sigma.text())
degree = int(self.degree.text())
self.axes.clear()
self.axes.grid(True)
self.axes.plot(self.data.x1_pos, self.data.x2_pos, 'ro')
self.axes.plot(self.data.x1_neg, self.data.x2_neg, 'bo')
# train svm
labels = self.data.get_labels()
print type(labels)
lab = BinaryLabels(labels)
features = self.data.get_examples()
train = RealFeatures(features)
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)
cost = float(self.cost.text())
print "cost", cost
svm = LibSVM(cost, gk, lab)
svm.train()
svm.set_epsilon(1e-2)
x, y, z = util.compute_output_plot_isolines(svm, gk, train)
plt=self.axes.pcolor(x, y, z, shading='interp')
CS=self.axes.contour(x, y, z, [-1,0,1], linewidths=1, colors='black', hold=True)
#CS=self.axes.contour(x, y, z, linewidths=1, colors='black', hold=True)
#CS=self.axes.contour(x, y, z, 5, linewidths=1, colors='black', hold=True)
matplotlib.pyplot.clabel(CS, inline=1, fontsize=10)
self.axes.set_xlim((-5,5))
self.axes.set_ylim((-5,5))
cmap = matplotlib.cm.jet
norm = mpl.colors.Normalize(numpy.min(z), numpy.max(z))
print CS.get_clim()
if not self.cax:
self.cax, kw = make_axes(self.axes)
# ColorbarBase derives from ScalarMappable and puts a colorbar
# in a specified axes, so it has everything needed for a
# standalone colorbar. There are many more kwargs, but the
# following gives a basic continuous colorbar with ticks
# and labels.
cb1 = mpl.colorbar.ColorbarBase(self.cax, cmap=cmap,
norm=norm)
self.canvas.draw()
def train_svm(self):
width = float(self.sigma.text())
degree = int(self.degree.text())
self.axes.clear()
self.axes.grid(True)
self.axes.plot(self.data.x1_pos, self.data.x2_pos, 'ro')
self.axes.plot(self.data.x1_neg, self.data.x2_neg, 'bo')
# train svm
labels = self.data.get_labels()
print type(labels)
lab = BinaryLabels(labels)
features = self.data.get_examples()
train = RealFeatures(features)
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)
cost = float(self.cost.text())
print "cost", cost
svm = LibSVM(cost, gk, lab)
svm.train()
svm.set_epsilon(1e-2)
x, y, z = util.compute_output_plot_isolines(svm, gk, train)
plt=self.axes.pcolor(x, y, z)
CS=self.axes.contour(x, y, z, [-1,0,1], linewidths=1, colors='black', hold=True)
#CS=self.axes.contour(x, y, z, linewidths=1, colors='black', hold=True)
#CS=self.axes.contour(x, y, z, 5, linewidths=1, colors='black', hold=True)
matplotlib.pyplot.clabel(CS, inline=1, fontsize=10)
self.axes.set_xlim((-5,5))
self.axes.set_ylim((-5,5))
cmap = matplotlib.cm.jet
norm = matplotlib.colors.Normalize(numpy.min(z), numpy.max(z))
print CS.get_clim()
if not self.cax:
self.cax, kw = make_axes(self.axes)
# ColorbarBase derives from ScalarMappable and puts a colorbar
# in a specified axes, so it has everything needed for a
# standalone colorbar. There are many more kwargs, but the
# following gives a basic continuous colorbar with ticks
# and labels.
cb1 = matplotlib.colorbar.ColorbarBase(self.cax, cmap=cmap,
norm=norm)
self.canvas.draw()
util.set_title('SVM')
util.NUM_EXAMPLES = 200
width = 5
# positive examples
pos = util.get_realdata(True)
pylab.plot(pos[0, :], pos[1, :], "rs")
# negative examples
neg = util.get_realdata(False)
pylab.plot(neg[0, :], neg[1, :], "bo")
# train svm
labels = util.get_labels()
train = util.get_realfeatures(pos, neg)
gk = GaussianKernel(train, train, width)
svm = LibSVM(10.0, gk, labels)
svm.train()
x, y, z = util.compute_output_plot_isolines(svm, gk, train)
pylab.contour(x, y, z, linewidths=1, colors='black', hold=True)
pylab.axis('tight')
pylab.title('Binary SVM classification with Gaussian kernel')
pylab.xlabel('x')
pylab.ylabel('y')
pylab.connect('key_press_event', util.quit)
pylab.show()
dense=util.get_realfeatures(pos, neg)
train=SparseRealFeatures()
train.obtain_from_simple(dense)
svm=SVMLin(C, train, labels)
svm.train()
lk=LinearKernel(dense, dense)
try:
svmlight=LibSVM(C, lk, labels)
except NameError:
print 'No SVMLight support available'
import sys
sys.exit(1)
svmlight.train()
x, y, z=util.compute_output_plot_isolines(svm, None, None, True, pos, neg)
x, y, zlight=util.compute_output_plot_isolines(svmlight, lk, dense, False, pos, neg)
c=pcolor(x, y, z, shading='interp')
contour(x, y, z, linewidths=1, colors='black', hold=True)
colorbar(c)
scatter(pos[0,:], pos[1,:],s=20, c='r')
scatter(neg[0,:], neg[1,:],s=20, c='b')
axis('tight')
connect('key_press_event', util.quit)
figure()
util.set_title('SVM Linear 2')
c=pcolor(x, y, zlight, shading='interp')
contour(x, y, zlight, linewidths=1, colors='black', hold=True)
colorbar(c)
pos = util.get_realdata(True)
neg = util.get_realdata(False)
traindatList[i] = concatenate((pos, neg), axis=1)
trainfeatList[i] = util.get_realfeatures(pos, neg)
trainlabsList[i] = util.get_labels(True)
trainlabList[i] = util.get_labels()
kernelList[i] = GaussianKernel(trainfeatList[i], trainfeatList[i], width)
svmList[i] = LibSVM(10, kernelList[i], trainlabList[i])
for i in range(num_svms):
print "Training svm nr. %d" % (i)
currentSVM = svmList[i]
currentSVM.train()
print currentSVM.get_num_support_vectors()
print "Done."
x, y, z = util.compute_output_plot_isolines(currentSVM, kernelList[i],
trainfeatList[i])
subplot(num_svms / 2, 2, i + 1)
pcolor(x, y, z)
contour(x, y, z, linewidths=1, colors='black', hold=True)
scatter(traindatList[i][0, :],
traindatList[i][1, :],
s=20,
marker='o',
c=trainlabsList[i],
hold=True)
axis('tight')
connect('key_press_event', util.quit)
show()
import util
util.set_title('SVM')
util.NUM_EXAMPLES=200
width=5
# positive examples
pos=util.get_realdata(True)
plot(pos[0,:], pos[1,:], "r.")
# negative examples
neg=util.get_realdata(False)
plot(neg[0,:], neg[1,:], "b.")
# train svm
labels=util.get_labels()
train=util.get_realfeatures(pos, neg)
gk=GaussianKernel(train, train, width)
svm = LibSVM(10.0, gk, labels)
svm.train()
x, y, z=util.compute_output_plot_isolines(svm, gk, train)
pcolor(x, y, z, shading='interp')
contour(x, y, z, linewidths=1, colors='black', hold=True)
axis('tight')
connect('key_press_event', util.quit)
show()
import util
util.set_title('LDA')
util.DISTANCE = 0.5
gamma = 0.1
# positive examples
pos = util.get_realdata(True)
plot(pos[0, :], pos[1, :], "r.")
# negative examples
neg = util.get_realdata(False)
plot(neg[0, :], neg[1, :], "b.")
# train lda
labels = util.get_labels()
features = util.get_realfeatures(pos, neg)
lda = LDA(gamma, features, labels)
lda.train()
# compute output plot iso-lines
x, y, z = util.compute_output_plot_isolines(lda)
c = pcolor(x, y, z)
contour(x, y, z, linewidths=1, colors='black', hold=True)
colorbar(c)
connect('key_press_event', util.quit)
show()
trainlabList = [None]*num_svms
trainlabsList = [None]*num_svms
kernelList = [None]*num_svms
for i in range(num_svms):
pos=util.get_realdata(True)
neg=util.get_realdata(False)
traindatList[i] = concatenate((pos, neg), axis=1)
trainfeatList[i] = util.get_realfeatures(pos, neg)
trainlabsList[i] = util.get_labels(True)
trainlabList[i] = util.get_labels()
kernelList[i] = GaussianKernel(trainfeatList[i], trainfeatList[i], width)
svmList[i] = LibSVM(10, kernelList[i], trainlabList[i])
for i in range(num_svms):
print "Training svm nr. %d" % (i)
currentSVM = svmList[i]
currentSVM.train()
print currentSVM.get_num_support_vectors()
print "Done."
x, y, z=util.compute_output_plot_isolines(
currentSVM, kernelList[i], trainfeatList[i])
subplot(num_svms/2, 2, i+1)
pcolor(x, y, z, shading='interp')
contour(x, y, z, linewidths=1, colors='black', hold=True)
scatter(traindatList[i][0,:],traindatList[i][1,:], s=20, marker='o', c=trainlabsList[i], hold=True)
axis('tight')
connect('key_press_event', util.quit)
show()
util.set_title('KernelRidgeRegression')
width = 20
# positive examples
pos = util.get_realdata(True)
plot(pos[0, :], pos[1, :], "r.")
# negative examples
neg = util.get_realdata(False)
plot(neg[0, :], neg[1, :], "b.")
# train krr
labels = util.get_labels(type='regression')
train = util.get_realfeatures(pos, neg)
gk = GaussianKernel(train, train, width)
krr = KernelRidgeRegression()
krr.set_labels(labels)
krr.set_kernel(gk)
krr.set_tau(1e-3)
krr.train()
# compute output plot iso-lines
x, y, z = util.compute_output_plot_isolines(krr, gk, train, regression=True)
pcolor(x, y, z)
contour(x, y, z, linewidths=1, colors='black', hold=True)
connect('key_press_event', util.quit)
show()
width=20
# positive examples
pos=util.get_realdata(True)
plot(pos[0,:], pos[1,:], "r.")
# negative examples
neg=util.get_realdata(False)
plot(neg[0,:], neg[1,:], "b.")
# train svm
labels = util.get_labels(type='regression')
train = util.get_realfeatures(pos, neg)
gk=GaussianKernel(train, train, width)
krr = KernelRidgeRegression()
krr.set_labels(labels)
krr.set_kernel(gk)
krr.set_tau(1e-3)
krr.train()
# compute output plot iso-lines
x, y, z=util.compute_output_plot_isolines(krr, gk, train, regression=True)
pcolor(x, y, z, shading='interp')
contour(x, y, z, linewidths=1, colors='black', hold=True)
connect('key_press_event', util.quit)
show()
util.set_title('LDA')
util.DISTANCE=0.5
gamma=0.1
# positive examples
pos=util.get_realdata(True)
plot(pos[0,:], pos[1,:], "r.")
# negative examples
neg=util.get_realdata(False)
plot(neg[0,:], neg[1,:], "b.")
# train lda
labels=util.get_labels()
features=util.get_realfeatures(pos, neg)
lda=LDA(gamma, features, labels)
lda.train()
# compute output plot iso-lines
x, y, z=util.compute_output_plot_isolines(lda)
c=pcolor(x, y, z, shading='interp')
contour(x, y, z, linewidths=1, colors='black', hold=True)
colorbar(c)
connect('key_press_event', util.quit)
show()
