def __init__(self, DATASET):
self.DATA = DATASET
self.Class1_train_Matrix = sf.get_Matrix(DATASET[0])
self.Class2_train_Matrix = sf.get_Matrix(DATASET[1])
self.Class3_train_Matrix = sf.get_Matrix(DATASET[2])
for i in range(len(DATASET)):
self.mew.append(sf.Mean(DATASET[i]))
def plot_model(self):
self.Class1_train_Matrix = sf.get_Matrix(self.DATA[0])
self.Class2_train_Matrix = sf.get_Matrix(self.DATA[1])
self.Class3_train_Matrix = sf.get_Matrix(self.DATA[2])
sf.plot_fourth(self.class1, self.class2, self.class3, self.DATA,
self.Class1_train_Matrix, self.Class2_train_Matrix,
self.Class3_train_Matrix, self.mew)
def __init__(self, DATASET, RANGE):
self.DATA = DATASET
self.RANGE = RANGE
Class1_train_Matrix = sf.get_Matrix(DATASET[0])
Class2_train_Matrix = sf.get_Matrix(DATASET[1])
Class3_train_Matrix = sf.get_Matrix(DATASET[2])
for i in range(len(DATASET)):
self.mew.append(sf.Mean(DATASET[i]))
for i in range(2):
for j in range(2):
self.Matrix[i][j] = Class1_train_Matrix[i][
j] + Class2_train_Matrix[i][j] + Class3_train_Matrix[i][j]
for i in range(2):
for j in range(2):
self.Matrix[i][j] = self.Matrix[i][j] / 3
def __init__(self, DATASET,RANGE): self.DATA=DATASET self.RANGE=RANGE Class1_train_Matrix=sf.get_Matrix(DATASET[0]) print(Class1_train_Matrix) Class2_train_Matrix=sf.get_Matrix(DATASET[1]) Class3_train_Matrix=sf.get_Matrix(DATASET[2]) self.mew.append(sf.Mean(DATASET[0])) self.mew.append(sf.Mean(DATASET[1])) self.mew.append(sf.Mean(DATASET[2])) for i in range(2): for j in range(2): self.Matrix[i][j]=Class1_train_Matrix[i][j]+Class2_train_Matrix[i][j]+Class3_train_Matrix[i][j] for i in range(2): for j in range(2): self.Matrix[i][j]=self.Matrix[i][j]/3 temp=(self.Matrix[0][0]+self.Matrix[1][1])/2 self.Matrix[0][0]=temp self.var=temp self.Matrix[1][1]=temp self.Matrix[0][1]=0 self.Matrix[1][0]=0
def get_ConfMatrix(self, TESTSET):
CONF = [[0, 0, 0], [0, 0, 0], [0, 0, 0]]
self.Class1_test_Matrix = sf.get_Matrix(TESTSET[0])
self.Class2_test_Matrix = sf.get_Matrix(TESTSET[1])
self.Class3_test_Matrix = sf.get_Matrix(TESTSET[2])
self.mew = []
for i in range(len(TESTSET)):
self.mew.append(sf.Mean(TESTSET[i]))
# print (self.Class1_test_Matrix,self.Class2_test_Matrix)
inv_class1 = sf.get_Inverse(self.Class1_test_Matrix)
inv_class2 = sf.get_Inverse(self.Class2_test_Matrix)
inv_class3 = sf.get_Inverse(self.Class3_test_Matrix)
temp1 = sf.get_Matrix(TESTSET[0])
temp2 = sf.get_Matrix(TESTSET[1])
temp3 = sf.get_Matrix(TESTSET[2])
# print (temp1,temp2)
for i in range(len(TESTSET)):
for j in range(len(TESTSET[i])):
temp = [0, 0, 0]
val1 = self.getGx(TESTSET[i][j][0], TESTSET[i][j][1],
inv_class1, self.mew[0], temp1,
len(TESTSET[0]))
val2 = self.getGx(TESTSET[i][j][0], TESTSET[i][j][1],
inv_class2, self.mew[1], temp2,
len(TESTSET[1]))
val3 = self.getGx(TESTSET[i][j][0], TESTSET[i][j][1],
inv_class3, self.mew[2], temp3,
len(TESTSET[2]))
if (max(val1, val2, val3) == val1): CONF[i][0] = CONF[i][0] + 1
elif (max(val1, val2, val3) == val2):
CONF[i][1] = CONF[i][1] + 1
else:
CONF[i][2] = CONF[i][2] + 1
for i in range(3):
for j in range(3):
print(CONF[i][j], end=" ")
print("")
sf.get_Score(CONF)
def get_pair(self, data_id, DATASET):
step = 1
left_margin, right_margin, top_margin, bottom_margin = 0, 0, 0, 0
if (data_id == 1):
step = 0.2
left_margin, right_margin, top_margin, bottom_margin = -10, 25, -15, 20
elif (data_id == 2):
step = 0.05
left_margin, right_margin, top_margin, bottom_margin = -3, 3, -3, 3
elif (data_id == 3):
step = 20
left_margin, right_margin, top_margin, bottom_margin = -500, 2100, 0, 3000
self.Class1_train_Matrix = sf.get_Matrix(DATASET[0])
self.Class2_train_Matrix = sf.get_Matrix(DATASET[1])
self.Class3_train_Matrix = sf.get_Matrix(DATASET[2])
inv_class1 = sf.get_Inverse(self.Class1_train_Matrix)
inv_class2 = sf.get_Inverse(self.Class2_train_Matrix)
inv_class3 = sf.get_Inverse(self.Class3_train_Matrix)
temp1 = sf.get_Matrix(DATASET[0])
temp2 = sf.get_Matrix(DATASET[1])
temp3 = sf.get_Matrix(DATASET[2])
classA, classB = [], []
i = left_margin
while (i < right_margin + 1):
j = top_margin
while (j < bottom_margin + 1):
val1 = self.getGx(i, j, inv_class1, self.mew[0],
self.Class1_train_Matrix, len(self.DATA[0]))
val2 = self.getGx(i, j, inv_class2, self.mew[1],
self.Class2_train_Matrix, len(self.DATA[1]))
if (val1 > val2): classA.append([i, j])
else: classB.append([i, j])
j += step
i += step
# sf.plot_fourth_pair(classA, classB,1,2,self.DATA,self.Class1_train_Matrix,self.Class2_train_Matrix,self.mew)
sf.plot_fourth_pair(classA, classB, 1, 2, self.DATA, temp1, temp2,
self.mew)
classB, classC = [], []
i = left_margin
while (i < right_margin + 1):
j = top_margin
while (j < bottom_margin + 1):
val2 = self.getGx(i, j, inv_class2, self.mew[1],
self.Class2_train_Matrix, len(self.DATA[1]))
val3 = self.getGx(i, j, inv_class3, self.mew[2],
self.Class3_train_Matrix, len(self.DATA[2]))
if (val3 > val2): classC.append([i, j])
else: classB.append([i, j])
j += step
i += step
sf.plot_fourth_pair(classB, classC, 2, 3, self.DATA, temp2, temp3,
self.mew)
classA, classC = [], []
i = left_margin
while (i < right_margin + 1):
j = top_margin
while (j < bottom_margin + 1):
val1 = self.getGx(i, j, inv_class1, self.mew[0],
self.Class1_train_Matrix, len(self.DATA[0]))
val3 = self.getGx(i, j, inv_class3, self.mew[2],
self.Class3_train_Matrix, len(self.DATA[2]))
if (val3 > val1): classC.append([i, j])
else: classA.append([i, j])
j += step
i += step
sf.plot_fourth_pair(classA, classC, 1, 3, self.DATA, temp1, temp3,
self.mew)
return
def get_ConfMatrix_pair(self, TESTSET):
CONF = [[0, 0], [0, 0]]
temp1 = sf.get_Matrix(TESTSET[0])
temp2 = sf.get_Matrix(TESTSET[1])
temp3 = sf.get_Matrix(TESTSET[2])
self.mew = []
for i in range(len(TESTSET)):
self.mew.append(sf.Mean(TESTSET[i]))
inv_class1 = sf.get_Inverse(temp1)
inv_class2 = sf.get_Inverse(temp2)
inv_class3 = sf.get_Inverse(temp3)
self.Class1_test_Matrix = sf.get_Matrix(TESTSET[0])
self.Class2_test_Matrix = sf.get_Matrix(TESTSET[1])
self.Class3_test_Matrix = sf.get_Matrix(TESTSET[2])
# for i in range(len(TESTSET)):
case = [0, 1]
for i in case:
for j in range(len(TESTSET[i])):
temp = [0, 0, 0]
val1 = self.getGx(TESTSET[i][j][0], TESTSET[i][j][1],
inv_class1, self.mew[0],
self.Class1_test_Matrix, len(TESTSET[0]))
val2 = self.getGx(TESTSET[i][j][0], TESTSET[i][j][1],
inv_class2, self.mew[1],
self.Class2_test_Matrix, len(TESTSET[1]))
if (val1 > val2): CONF[i][0] = CONF[i][0] + 1
else: CONF[i][1] = CONF[i][1] + 1
print("Confusion Matrix for class 1 & 2")
for i in range(2):
for j in range(2):
print(CONF[i][j], end=" ")
print("")
sf.get_Score(CONF)
CONF = [[0, 0], [0, 0]]
case = [1, 2]
for i in case:
for j in range(len(TESTSET[i])):
temp = [0, 0, 0]
val2 = self.getGx(TESTSET[i][j][0], TESTSET[i][j][1],
inv_class2, self.mew[1],
self.Class2_test_Matrix, len(TESTSET[1]))
val3 = self.getGx(TESTSET[i][j][0], TESTSET[i][j][1],
inv_class3, self.mew[2],
self.Class3_test_Matrix, len(TESTSET[2]))
if (val2 > val3): CONF[i - 1][0] = CONF[i - 1][0] + 1
else: CONF[i - 1][1] = CONF[i - 1][1] + 1
print("Confusion Matrix for class 2 & 3")
for i in range(2):
for j in range(2):
print(CONF[i][j], end=" ")
print("")
sf.get_Score(CONF)
CONF = [[0, 0], [0, 0]]
case = [0, 2]
for i in case:
for j in range(len(TESTSET[i])):
temp = [0, 0, 0]
val1 = self.getGx(TESTSET[i][j][0], TESTSET[i][j][1],
inv_class1, self.mew[0],
self.Class1_test_Matrix, len(TESTSET[0]))
val3 = self.getGx(TESTSET[i][j][0], TESTSET[i][j][1],
inv_class3, self.mew[2],
self.Class3_test_Matrix, len(TESTSET[2]))
k = 0
if (i == 2): k = 1
if (val1 > val3): CONF[k][0] = CONF[k][0] + 1
else: CONF[k][1] = CONF[k][1] + 1
print("Confusion Matrix for class 2 & 3")
for i in range(2):
for j in range(2):
print(CONF[i][j], end=" ")
print("")
sf.get_Score(CONF)