def testRbf(k1=1.3):
dataArr, labelArr = SMO.loadDataSet('testSetRBF.txt')
b, alphas = smoP(dataArr, labelArr, 200, 0.0001, 10000, ('rbf', k1))
dataMat = mat(dataArr)
labelMat = mat(labelArr).transpose()
svInd = nonzero(alphas.A > 0)[0]
sVs = dataMat[svInd]
labelLSV = labelMat[svInd]
print("there are %d Support Vectors" % shape(sVs)[0])
m, n = shape(dataMat)
errorCount = 0
for i in range(m):
kernelEval = kernelTrans(sVs, dataMat[i, :], ('rbf', k1))
predict = kernelEval.T * multiply(labelLSV, alphas[svInd]) + b
if sign(predict) != sign(labelArr[i]): errorCount += 1
print("the training error rate is: %f" % (float(errorCount) / m))
dataArr, labelArr = SMO.loadDataSet('testSetRBF2.txt')
errorCount = 0
dataMat = mat(dataArr)
labelMat = mat(labelArr).transpose()
m, n = shape(dataMat)
for i in range(m):
kernelEval = kernelTrans(sVs, dataMat[i, :], ('rbf', k1))
predict = kernelEval.T * multiply(labelLSV, alphas[svInd]) + b
if sign(predict) != sign(labelArr[i]): errorCount += 1
print("the test error rate is: %f" % (float(errorCount) / m))
class TestSMOOnePoint(unittest.TestCase):
def setUp(self):
X = np.array([[1,1]])
Y = np.array([1])
self.SMOobj = SMO(X, Y, 0.1, 0.001, 0.001)
self.SMOobj.SMO_main()
def test_weight_correctness(self):
self.assertEqual(self.SMOobj.get_weight(), np.array([0.0, 0.0]))
def test_threshold_correctness(self):
self.assertEqual(self.SMOobj.get_threshold(), 0.0)
def selectJ(i, oS, Ei):
maxK = -1
maxDeltaE = 0
Ej = 0
#set corresponding cached value as valid
oS.eCache[i] = [1, Ei]
#return the corresponding alpha of non-zero E
validEcacheList = nonzero(oS.eCache[:, 0].A)[0]
#select the alpha j that maximum abs(Ei - Ek)
if (len(validEcacheList)) > 1:
for k in validEcacheList:
if k == i:
continue
Ek = calcEk(oS, k)
deltaE = abs(Ei - Ek)
if (deltaE > maxDeltaE):
maxK = k
maxDeltaE = deltaE
Ej = Ek
return maxK, Ej
#if there is no such j, random select a j
else:
j = smo.selectJrand(i, oS.m)
Ej = calcEk(oS, j)
return j, Ej
def selectJ(i, oS, Ei):
maxK = -1
maxDeltaE = 0
Ej = 0
#set corresponding cached value as valid
oS.eCache[i] = [1,Ei]
#return the corresponding alpha of non-zero E
validEcacheList = nonzero(oS.eCache[:,0].A)[0]
#select the alpha j that maximum abs(Ei - Ek)
if(len(validEcacheList)) > 1:
for k in validEcacheList:
if k == i:
continue
Ek = calcEk(oS, k)
deltaE = abs(Ei - Ek)
if(deltaE > maxDeltaE):
maxK = k; maxDeltaE = deltaE; Ej = Ek
return maxK, Ej
#if there is no such j, random select a j
else:
j = smo.selectJrand(i, oS.m)
Ej = calcEk(oS,j)
return j, Ej
def selector(algo,func_details,popSize,Iter,succ_rate,mean_feval):
function_name=func_details[0]
lb=func_details[1]
ub=func_details[2]
dim=func_details[3]
acc_err=func_details[4]
obj_val=func_details[5]
if(algo==0):
x,succ_rate,mean_feval=smo.main(getattr(benchmarks, function_name),lb,ub,dim,popSize,Iter,acc_err,obj_val,succ_rate,mean_feval)
return x,succ_rate,mean_feval
def innerL(i, oS):
Ei = calcEk(oS, i)
if((oS.labelMat[i]*Ei < -oS.tol) and (oS.alphas[i] < oS.C)) or \
( (oS.labelMat[i]*Ei > oS.tol) and (oS.alphas[i] > 0)):
#improve from simple SMP
j, Ej = selectJ(i, oS, Ei)
alphaIold = oS.alphas[i].copy()
alphaJold = oS.alphas[j].copy()
if (oS.labelMat[i] != oS.labelMat[j]):
L = max(0, oS.alphas[j] - oS.alphas[i])
H = min(oS.C, oS.C + oS.alphas[j] - oS.alphas[i])
else:
L = max(0, oS.alphas[j] + oS.alphas[i] - oS.C)
H = min(oS.C, oS.alphas[j] + oS.alphas[i])
if L == H:
#print "L==H"
return 0
eta = 2.0 * oS.X[i, :] * oS.X[j, :].T - oS.X[i, :] * oS.X[
i, :].T - oS.X[j, :] * oS.X[j, :].T
if eta >= 0:
#print "eta>=0"
return 0
oS.alphas[j] -= oS.labelMat[j] * (Ei - Ej) / eta
oS.alphas[j] = smo.clipAlpha(oS.alphas[j], H, L)
updateEk(oS, j)
if (abs(oS.alphas[j] - alphaJold) < 0.00001):
#print "j not moving enough"
return 0
oS.alphas[i] += oS.labelMat[j] * oS.labelMat[i] * (alphaJold -
oS.alphas[j])
updateEk(oS, i)
#update constant b
b1 = oS.b - Ei - oS.labelMat[i] * (oS.alphas[i] - alphaIold) * oS.X[
i, :] * oS.X[i, :].T - oS.labelMat[j] * (
oS.alphas[j] - alphaJold) * oS.X[i, :] * oS.X[j, :].T
b2 = oS.b - Ej - oS.labelMat[i] * (oS.alphas[i] - alphaIold) * oS.X[
i, :] * oS.X[j, :].T - oS.labelMat[j] * (
oS.alphas[j] - alphaJold) * oS.X[j, :] * oS.X[j, :].T
if (0 < oS.alphas[i]) and (oS.C > oS.alphas[i]): oS.b = b1
elif (0 < oS.alphas[j]) and (oS.C > oS.alphas[j]): oS.b = b2
else: oS.b = (b1 + b2) / 2.0
return 1
else:
return 0
def selector(algo,func_details,popSize,Iter,succ_rate,mean_feval):
function_name=func_details[0]
lb=func_details[1]
ub=func_details[2]
dim=func_details[3]
acc_err=func_details[4]
obj_val=func_details[5]
#selection of different parameters
if(algo==0):
x,succ_rate,mean_feval=smo.main(getattr(benchmarks, function_name),lb,ub,dim,popSize,Iter,acc_err,obj_val,succ_rate,mean_feval) #getting attributes from different file
return x,succ_rate,mean_feval
def testRbf(k1=1.3):
#load data
dataArr, labelArr = smo.loadDataSet('testSetRBF.txt')
#training
b, alphas = smoP(dataArr, labelArr, 200, 0.0001, 10000, ('rbf',k1))
datMat = mat(dataArr)
labelMat = mat(labelArr).transpose()
#build support vector
svInd = nonzero(alphas.A > 0)[0]
sVs = datMat[svInd]
labelSV = labelMat[svInd]
print "there are %d Support Vectors" % shape(sVs)[0]
#training error rate
m,n = shape(datMat)
errorCount = 0
for i in range(m):
kernelEval = kernelTrans(sVs, datMat[i,:],('rbf',k1))
predict = kernelEval.T * multiply(labelSV, alphas[svInd]) + b
if sign(predict) != sign(labelArr[i]):
errorCount += 1
print "the test error rate is: %f" %(float(errorCount)/m)
#test error rate
dataArr, labelArr = smo.loadDataSet('testSetRBF2.txt')
errorCount = 0
datMat = mat(dataArr)
labelmat = mat(labelArr).transpose()
m,n = shape(datMat)
for i in range(m):
kernelEval = kernelTrans(sVs, datMat[i,:],('rbf',k1))
predict = kernelEval.T * multiply(labelSV, alphas[svInd]) + b
if sign(predict) != sign(labelArr[i]):
errorCount += 1
print "the test error rate is: %f" %(float(errorCount)/m)
def innerL(i, oS):
Ei = calcEk(oS,i)
if((oS.labelMat[i]*Ei < -oS.tol) and (oS.alphas[i] < oS.C)) or \
( (oS.labelMat[i]*Ei > oS.tol) and (oS.alphas[i] > 0)):
#improve from simple SMP
j, Ej = selectJ(i, oS, Ei)
alphaIold = oS.alphas[i].copy()
alphaJold = oS.alphas[j].copy()
if(oS.labelMat[i] != oS.labelMat[j]):
L = max(0, oS.alphas[j] - oS.alphas[i])
H = min(oS.C, oS.C + oS.alphas[j] - oS.alphas[i])
else:
L = max(0, oS.alphas[j] + oS.alphas[i] - oS.C)
H = min(oS.C, oS.alphas[j] + oS.alphas[i])
if L==H:
#print "L==H"
return 0
#Code different from plattSMO.py
eta = 2.0*oS.K[i,j] - oS.K[i,i] - oS.K[j,j]
if eta >= 0:
#print "eta>=0"
return 0
oS.alphas[j] -= oS.labelMat[j]*(Ei - Ej)/eta
oS.alphas[j] = smo.clipAlpha(oS.alphas[j], H, L)
updateEk(oS,j)
if(abs(oS.alphas[j] - alphaJold) < 0.00001):
#print "j not moving enough"
return 0
oS.alphas[i] += oS.labelMat[j]*oS.labelMat[i]*(alphaJold - oS.alphas[j])
updateEk(oS,i)
#update constant b
b1 = oS.b - Ei- oS.labelMat[i]*(oS.alphas[i]-alphaIold)*oS.K[i,i] - oS.labelMat[j]*(oS.alphas[j]-
alphaJold)*oS.K[i,j]
b2 = oS.b - Ej- oS.labelMat[i]*(oS.alphas[i]-alphaIold)*oS.K[i,j] - oS.labelMat[j]*(oS.alphas[j]-
alphaJold)*oS.K[j,j]
if(0 < oS.alphas[i]) and (oS.C > oS.alphas[i]): oS.b = b1
elif (0 < oS.alphas[j]) and (oS.C > oS.alphas[j]): oS.b = b2
else: oS.b = (b1+b2)/2.0
return 1
else: return 0
def compute_alpha(self, X, y):
"""
n_samples, n_features = X.shape
K = self.kernel_matrix(X)
# Solves
# min 1/2 x^T P x + q^T x
# s.t.
# Gx \coneleq h
# Ax = b
P = cvxopt.matrix(np.outer(y, y) * K)
q = cvxopt.matrix(-1 * np.ones(n_samples))
# -a_i \leq 0
# TODO(tulloch) - modify G, h so that we have a soft-margin classifier
G_std = cvxopt.matrix(np.diag(np.ones(n_samples) * -1))
h_std = cvxopt.matrix(np.zeros(n_samples))
# a_i \leq c
G_slack = cvxopt.matrix(np.diag(np.ones(n_samples)))
h_slack = cvxopt.matrix(np.ones(n_samples) * self._c)
G = cvxopt.matrix(np.vstack((G_std, G_slack)))
h = cvxopt.matrix(np.vstack((h_std, h_slack)))
A = cvxopt.matrix(y, (1, n_samples))
b = cvxopt.matrix(0.0)
solution = cvxopt.solvers.qp(P, q, G, h, A, b)
# Lagrange multipliers
return np.ravel(solution['x'])
"""
SMOobj = SMO(X, y, self._c, self._tol, self._eps)
SMOobj.SMO_main()
return SMOobj.get_alpha()
def innerL(i, oS):
Ei = calcEk(oS, i)
if((oS.labelMat[i]*Ei<-oS.tol) and (oS.alphas[i]<oS.C)) or \
((oS.labelMat[i]*Ei>oS.tol) and (oS.alphas[i]>0)):
j, Ej = selectJ(i, oS, Ei) #选择alphaJ
alphaIold = oS.alphas[i].copy()
alphaJold = oS.alphas[j].copy()
#对最有值的边界确定
if (oS.labelMat[i] != oS.labelMat[j]):
L = max(0, oS.alphas[j] - oS.alphas[i])
H = min(oS.C, oS.C + oS.alphas[j] - oS.alphas[i])
else:
L = max(0, oS.alphas[j] + oS.alphas[i] - oS.C)
H = min(oS.C, oS.alphas[j] + oS.alphas[i])
if L == H:
print("L==H")
return 0
#eta = 2.0*oS.X[i,:]*oS.X[j,:].T-oS.X[i,:]*oS.X[i,:].T- oS.X[j,:]*oS.X[j,:].T
eta = 2.0 * oS.K[i, j] - oS.K[i, i] - oS.K[j, j] #<统计学习方法>Page128
if eta >= 0:
print("eta>=0")
return 0
oS.alphas[j] -= oS.labelMat[j] * (Ei - Ej) / eta
oS.alphas[j] = SMO.clipAlpha(oS.alphas[j], H,
L) #将alphaJ限定在[L,H]之间 更新alphas[j]
updateEk(oS, j)
if (abs(oS.alphas[j] - alphaJold) < 0.00001):
print("j not moving enough!")
return 0
oS.alphas[i] += oS.labelMat[j] * oS.labelMat[i] * (
alphaJold - oS.alphas[j]) #更新alpha[i]
updateEk(oS, i)
b1 = oS.b - Ei - oS.labelMat[i] * (oS.alphas[i] - alphaIold) * oS.K[
i, i] - oS.labelMat[j] * (oS.alphas[j] - alphaJold) * oS.K[i, j]
b2 = oS.b - Ej - oS.labelMat[i] * (oS.alphas[i] - alphaIold) * oS.K[
i, j] - oS.labelMat[j] * (oS.alphas[j] - alphaJold) * oS.K[j, j]
# b1=oS.b-Ei-oS.labelMat[i]*(oS.alphas[i]-alphaIold)*oS.X[i,:]*oS.X[i,:].T-oS.labelMat[j]*\
# (oS.alphas[j]-alphaJold)*oS.X[i,:]*oS.X[j,:].T
# b2=oS.b-Ej-oS.labelMat[i]*(oS.alphas[i]-alphaIold)*oS.X[i,:]*oS.X[j,:].T-oS.labelMat[j]*\
# (oS.alphas[j]-alphaJold)*oS.X[j,:]*oS.X[j,:].T
if (0 < oS.alphas[i]) and (oS.C > oS.alphas[i]): oS.b = b1
elif (0 < oS.alphas[j]) and (oS.C > oS.alphas[j]): oS.b = b2
else: oS.b = (b1 + b2) / 2.0
return 1
else:
return 0
class TestSMOTwoPoints(unittest.TestCase):
def setUp(self):
X = np.array([[-1,-1], [1,1]])
Y = np.array([-1, 1])
self.SMOobj = SMO(X, Y, 0.1, 0.001, 0.001)
def test_weight_correctness(self):
self.assertTrue(self.SMOobj.get_weight()[0]/self.SMOobj.get_weight()[1] >= 1 - 1e-5
and self.SMOobj.get_weight()[0]/self.SMOobj.get_weight()[1] <= 1 + 1e-5)
def test_threshold_correctness(self):
self.assertTrue(self.SMOobj.get_threshold() <= 1e-5 and self.SMOobj.get_threshold() >= -1e-5)
def draw(data, classes, sv, alphas, b, resolution):
mycm = mpl.cm.get_cmap('Paired')
one_min, one_max = data[:, 0].min()-0.1, data[:, 0].max()+0.1
two_min, two_max = data[:, 1].min()-0.1, data[:, 1].max()+0.1
xx1, xx2 = np.meshgrid(np.arange(one_min, one_max, (one_max-one_min)/resolution),
np.arange(two_min, two_max, (two_max-two_min)/resolution))
inputs = np.c_[xx1.ravel(), xx2.ravel()]
z = []
for i in range(len(inputs)):
z.append(SMO.predict(data, classes, alphas, b, inputs[i]))
result = np.array(z).reshape(xx1.shape)
plt.contourf(xx1, xx2, result, cmap=mycm)
plt.scatter(data[:, 0], data[:, 1], s=50, c=classes, cmap=mycm)
plt.scatter(sv[:,0], sv[:,1], s=10)
plt.show()
def selectJ(i, oS, Ei):
maxK = -1
maxDeltaE = 0
Ej = 0
oS.eCache[i] = [1, Ei] #eCahe是所有误差的缓存区
vaildEcacheList = nonzero(oS.eCache[:, 0].A)[0]
#.A表示转化为array,注意nonzero返回的成对的值,这里validEcache返回的是第一列非零的行索引.见博客:https://blog.csdn.net/qq_28773183/article/details/81013226
if (len(vaildEcacheList)) > 1:
for k in vaildEcacheList:
if k == i: continue
Ek = calcEk(oS, k) #计算Ek,有索引k即可
deltaE = abs(Ei - Ek)
if (deltaE > maxDeltaE):
maxK = k
maxDeltaE = deltaE
Ej = Ek
return maxK, Ej
else: #随机选择alphaJ
j = SMO.selectJrand(i, oS.m)
Ej = calcEk(oS, j)
return j, Ej
def main():
print '-----------------------load data and training-----------'
dataArr, labelArr = smo.loadDataSet('testSet.txt')
b, alphas = smoP(dataArr, labelArr, 0.6, 0.001, 40)
print "b", b
print "alphas > 0", alphas[alphas > 0]
#calculate weights
print '-----------------------calculate weights----------------'
ws = calcWs(alphas, dataArr, labelArr)
print "ws", ws
print '-----------------------test classify result----------------'
#process first node
datMat = mat(dataArr)
print "first node classify to: ", datMat[0]*mat(ws) + b
print "first node true label:", labelArr[0]
#process third node
print "third node classify to: ", datMat[2]*mat(ws) + b
print "third node true label: ", labelArr[2]
def main():
print '-----------------------load data and training-----------'
dataArr, labelArr = smo.loadDataSet('testSet.txt')
b, alphas = smoP(dataArr, labelArr, 0.6, 0.001, 40)
print "b", b
print "alphas > 0", alphas[alphas > 0]
#calculate weights
print '-----------------------calculate weights----------------'
ws = calcWs(alphas, dataArr, labelArr)
print "ws", ws
print '-----------------------test classify result----------------'
#process first node
datMat = mat(dataArr)
print "first node classify to: ", datMat[0] * mat(ws) + b
print "first node true label:", labelArr[0]
#process third node
print "third node classify to: ", datMat[2] * mat(ws) + b
print "third node true label: ", labelArr[2]
def train(self, C=[0.01, 1, 10, 100], tol=1e-3):
m = self.Y.shape[0]
A = [0] * 10
B = [0] * 10
indices = numpy.random.permutation(
self.X.shape[0]) # shape[0]表示第0轴的长度，通常是训练数据的数量
rand_data_x = self.X[indices]
rand_data_y = self.Y[indices] # data_y就是标记（label）
l = int(len(indices) / 10)
for i in range(9):
A[i] = rand_data_x[i * l:i * l + l]
B[i] = rand_data_y[i * l:i * l + l]
A[9] = rand_data_x[9 * l:]
B[9] = rand_data_y[9 * l:]
# '''
# X_num=self.X.shape[0]
# train_index=range(X_num)
# test_size=int(X_num*0.1)+1
# for i in range(9):
# test_index=[]
# for j in range(test_size):
# randomIndex=int(numpy.random.uniform(0,len(train_index)))
# test_index.append(train_index[randomIndex])
# #del train_index[randomIndex]
# A[i]=self.X[test_index,:]
# B[i]=self.Y[test_index,:]
# A[9]=self.X.ix_[train_index]
# B[9]=self.Y.ix_[train_index]
# '''
acc_best = 0
C_best = None
avg_acc = 0
# gamma_best = None
for CVal in C:
# for gammaVal in gamma:
# avg_acc = 0
for i in range(10):
X_test = A[i]
Y_test = B[i]
# X_train = None
# Y_train = None
#model= SMO.SMO_Model(X_train, Y_train, CVal, kernel,gammaVal, tol=1e-3, eps=1e-3)
#output_model=SMO.SMO(model)
#根据output_model的参数信息计算对应decision_function----->推得accuracy
#acc = _evaulate(output_model)
X_train = numpy.concatenate([
A[(i + 1) % 10], A[(i + 2) % 10], A[(i + 3) % 10],
A[(i + 4) % 10], A[(i + 5) % 10], A[(i + 6) % 10],
A[(i + 7) % 10], A[(i + 8) % 10], A[(i + 9) % 10]
],
axis=0)
Y_train = numpy.concatenate([
B[(i + 1) % 10], B[(i + 2) % 10], B[(i + 3) % 10],
B[(i + 4) % 10], B[(i + 5) % 10], B[(i + 6) % 10],
B[(i + 7) % 10], B[(i + 8) % 10], B[(i + 9) % 10]
],
axis=0)
# SMO.GG = gammaVal
# calculate Kernel Matrix then pass it to SMO.
if self.IK:
if self.kernel_dict['type'] == 'TANH':
K = Kernel.TANH(X_train.shape[0],
self.kernel_dict['c'],
self.kernel_dict['d'])
K.calculate(X_train)
elif self.kernel_dict['type'] == 'TL1':
K = Kernel.TL1(X_train.shape[0],
self.kernel_dict['rho'])
K.calculate(X_train)
p1, p2 = trans_mat(K.kernelMat)
K.kernelMat = np.dot((p1 - p2), K.kernelMat)
if self.kernel_dict['type'] == 'RBF':
K = Kernel.RBF(X_train.shape[0], self.kernel_dict['gamma'])
K.calculate(X_train)
elif self.kernel_dict['type'] == 'LINEAR':
K = Kernel.LINEAR(X_train.shape[0])
K.calculate(X_train)
elif self.kernel_dict['type'] == 'POLY':
K = Kernel.POLY(X_train.shape[0], self.kernel_dict['c'],
self.kernel_dict['d'])
K.calculate(X_train)
elif self.kernel_dict['type'] == 'TANH':
K = Kernel.TANH(X_train.shape[0], self.kernel_dict['c'],
self.kernel_dict['d'])
K.calculate(X_train)
elif self.kernel_dict['type'] == 'TL1':
K = Kernel.TL1(X_train.shape[0], self.kernel_dict['rho'])
K.calculate(X_train)
model = SMO.SMO_Model(X_train,
Y_train,
CVal,
K,
tol=1e-3,
eps=1e-3)
output_model = SMO.SMO(model)
#IK
if self.IK:
output_model.alphas = np.dot((p1 - p2),
output_model.alphas)
acc = SMO._evaluate(output_model, X_test, Y_test)
avg_acc = avg_acc + acc / 10
if avg_acc > acc_best:
acc_best = avg_acc
#更新C gamma
C_best = CVal
# gamma_best =gammaVal
# self.gamma = gamma_best
#最后一遍train
# SMO.GG = gamma_best
#!K
if self.IK:
if self.kernel_dict['type'] == 'TANH':
K = Kernel.TANH(self.X.shape[0], self.kernel_dict['c'],
self.kernel_dict['d'])
K.calculate(self.X)
elif self.kernel_dict['type'] == 'TL1':
K = Kernel.TL1(self.X.shape[0], self.kernel_dict['rho'])
K.calculate(self.X)
p1, p2 = trans_mat(K.kernelMat)
K.kernelMat = np.dot((p1 - p2), K.kernelMat)
if self.kernel_dict['type'] == 'RBF':
K = Kernel.RBF(self.X.shape[0], self.kernel_dict['gamma'])
K.calculate(self.X)
elif self.kernel_dict['type'] == 'LINEAR':
K = Kernel.LINEAR(self.X.shape[0])
K.calculate(self.X)
elif self.kernel_dict['type'] == 'POLY':
K = Kernel.POLY(self.X.shape[0], self.kernel_dict['c'],
self.kernel_dict['d'])
K.calculate(self.X)
elif self.kernel_dict['type'] == 'TANH':
K = Kernel.TANH(self.X.shape[0], self.kernel_dict['c'],
self.kernel_dict['d'])
K.calculate(self.X)
elif self.kernel_dict['type'] == 'TL1':
K = Kernel.TL1(self.X.shape[0], self.kernel_dict['rho'])
K.calculate(self.X)
SVM_model = SMO.SMO(
SMO.SMO_Model(self.X, self.Y, C_best, K, tol=1e-3, eps=1e-3))
# 参数传递给最后生成的SVM类
if self.IK:
SVM_model.alphas = np.dot((p1 - p2), SVM_model.alphas)
self.X = SVM_model.X
self.Y = SVM_model.y
self.kernel_dict = SVM_model.kernel
self.alphas = SVM_model.alphas
self.b = SVM_model.b
# C_best = C
# gamma_best =gamma
# (w,b) = SMO(X_train,Y_train,C_best,gamma_best,kernal,tol=1e-3)
# self.w = w
# self.b = b
return None
from SMO import *
from sklearn.model_selection import train_test_split
import numpy as np
x = np.random.normal(size=(5, ))
y = np.outer(x, x)
z = np.random.multivariate_normal(np.zeros(5) - 2, y, size=100)
x1 = np.random.normal(size=(5, ))
y1 = np.outer(x1, x1)
z1 = np.random.multivariate_normal(np.zeros(5) + 2, y1, size=100)
X_train = np.r_[z, z1]
Y_train = np.r_[[1 for x in range(100)], [-1 for x in range(100)]]
X_train, X_test, y_train, y_test = train_test_split(X_train,
Y_train,
test_size=0.33,
random_state=42)
model = SMO()
model.fit(X_train, y_train)
print('score:', model.score(X_test, y_test))
np.arange(two_min, two_max, (two_max-two_min)/resolution))
inputs = np.c_[xx1.ravel(), xx2.ravel()]
z = []
for i in range(len(inputs)):
z.append(SMO.predict(data, classes, alphas, b, inputs[i]))
result = np.array(z).reshape(xx1.shape)
plt.contourf(xx1, xx2, result, cmap=mycm)
plt.scatter(data[:, 0], data[:, 1], s=50, c=classes, cmap=mycm)
plt.scatter(sv[:,0], sv[:,1], s=10)
plt.show()
if __name__ == '__main__':
data_set, N = simulate_data()
data, classes = cook_data(data_set)
#print data
#print classes
alphas, b = SMO.run(data, classes)
print alphas
print b
sv = []
for i in range(len(alphas)):
if alphas[i] >= 0.001 and alphas[i] <= 100.0:
print alphas[i], data[i]
sv.append(data[i])
sv = np.array(sv)
print sv
draw(data, classes, sv, alphas, b, 100.)
def selector(algo, func_details, popSize, Iter, trainDataset, testDataset):
function_name = func_details[0]
lb = func_details[1]
ub = func_details[2]
DatasetSplitRatio = 2 / 3
dataTrain = "datasets/" + trainDataset
dataTest = "datasets/" + testDataset
Dataset_train = numpy.loadtxt(open(dataTrain, "rb"),
delimiter=",",
skiprows=0)
Dataset_test = numpy.loadtxt(open(dataTest, "rb"),
delimiter=",",
skiprows=0)
numRowsTrain = numpy.shape(Dataset_train)[
0] # number of instances in the train dataset
numInputsTrain = numpy.shape(
Dataset_train)[1] - 1 #number of features in the train dataset
numRowsTest = numpy.shape(Dataset_test)[
0] # number of instances in the test dataset
numInputsTest = numpy.shape(
Dataset_test)[1] - 1 #number of features in the test dataset
trainInput = Dataset_train[0:numRowsTrain, 0:-1]
trainOutput = Dataset_train[0:numRowsTrain, -1]
testInput = Dataset_test[0:numRowsTest, 0:-1]
testOutput = Dataset_test[0:numRowsTest, -1]
#number of hidden neurons
HiddenNeurons = numInputsTrain * 2 + 1
net = nl.net.newff([[0, 1]] * numInputsTrain, [HiddenNeurons, 1])
dim = (numInputsTrain * HiddenNeurons) + (2 * HiddenNeurons) + 1
if (algo == 0):
x = pso.PSO(getattr(costNN, function_name), lb, ub, dim, popSize, Iter,
trainInput, trainOutput, net)
if (algo == 1):
x = mvo.MVO(getattr(costNN, function_name), lb, ub, dim, popSize, Iter,
trainInput, trainOutput, net)
if (algo == 2):
x = gwo.GWO(getattr(costNN, function_name), lb, ub, dim, popSize, Iter,
trainInput, trainOutput, net)
if (algo == 3):
x = mfo.MFO(getattr(costNN, function_name), lb, ub, dim, popSize, Iter,
trainInput, trainOutput, net)
if (algo == 4):
x = cs.CS(getattr(costNN, function_name), lb, ub, dim, popSize, Iter,
trainInput, trainOutput, net)
if (algo == 5):
x = bat.BAT(getattr(costNN, function_name), lb, ub, dim, popSize, Iter,
trainInput, trainOutput, net)
if (algo == 6):
x = smo.main(getattr(costNN, function_name), lb, ub, dim, popSize,
Iter, 1.0e-5, trainInput, trainOutput, net)
# print(type(x),'------------------------------')
# Evaluate MLP classification model based on the training set
trainClassification_results = evalNet.evaluateNetClassifier(
x, trainInput, trainOutput, net)
x.trainAcc = trainClassification_results[0]
x.trainTP = trainClassification_results[1]
x.trainFN = trainClassification_results[2]
x.trainFP = trainClassification_results[3]
x.trainTN = trainClassification_results[4]
# Evaluate MLP classification model based on the testing set
testClassification_results = evalNet.evaluateNetClassifier(
x, testInput, testOutput, net)
x.testAcc = testClassification_results[0]
x.testTP = testClassification_results[1]
x.testFN = testClassification_results[2]
x.testFP = testClassification_results[3]
x.testTN = testClassification_results[4]
return x
calculate w and draw the picture,
the variable which the α not equal zero ,
we call support vector
'''
def calculateW(alphas, data, labels):
x = np.mat(data)
label = np.mat(labels).transpose()
m, n = np.shape(x)
w = np.zeros((n, 1))
for i in range(m):
w += np.multiply(alphas[i] * label[i], x[i, :].T)
return w
pass
if __name__ == '__main__':
data, label = Tool.loadDataSet('../Data/testSet.txt')
b, alphas = SMO.smo(data, label, kernel=False)
w = calculateW(alphas, data, label)
x = np.arange(0, 11)
print(w)
y = (-b - w[0] * x) / w[1]
Tool.drawDataset(data, label, x, y.tolist()[0], line=True, alphas=alphas)
data, label = Tool.loadDataSet('../Data/testSetRBF.txt')
b, alphas = SMO.smo(data, label, kernel=True, maxIter=100)
svInd = np.nonzero(alphas.A > 0)[0]
Tool.drawDataset(data, label, line=False, alphas=alphas)
def setUp(self):
X = np.array([[1,1]])
Y = np.array([1])
self.SMOobj = SMO(X, Y, 0.1, 0.001, 0.001)
self.SMOobj.SMO_main()
i.pop(0)
tmp = []
for j in range(784):
tmp.append(0)
for j in i:
plz = j.find(':')
tmp[int(j[0:plz])] = int(j[plz + 1:]) * r
dataset.append(tmp)
dataset = np.array(dataset)
label = np.array(label)
# ytmp = label.reshape(-1, 1) * 1.
# xtmp = ytmp * dataset
alpha, x, y, b = SMO(dataset, label, 10)
b = np.float(b)
temp = alpha * y
temp = np.transpose(temp)
omega = np.dot(temp, x)
# omega = np.transpose(omega)
datafile = open('test-01-images.svm', mode='r')
tedata = datafile.readlines()
datafile.close()
testdata = []
for i in tedata:
testdata.append(i.replace('\n', '').split(' ', i.count(' ')))
dataset = []
label = []
def setUp(self):
X = np.array([[-1,-1], [1,1]])
Y = np.array([-1, 1])
self.SMOobj = SMO(X, Y, 0.1, 0.001, 0.001)
