def visualize_isSatNode(domainSiz):
xAc1 = []
yAc1 = []
xAc2 = []
yAc2 = []
xAc3 = []
yAc3 = []
xAc4 = []
yAc4 = []
for node in range(1, 101, 20):
print(node)
xAc1.append(node)
xAc2.append(node)
xAc3.append(node)
xAc4.append(node)
timeAc1 = 0
timeAc2 = 0
timeAc3 = 0
timeAc4 = 0
for i in range(10):
csp = generateCSP(node, domainSiz)
g = csp[0]
constraints = csp[1]
domain = csp[2]
edge = csp[3]
domainSize = csp[4]
mn.executeAC1(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
mn.executeAC2(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
mn.executeAC3(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
mn.executeAC4(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
timeAc1 += mn.cnt1
timeAc2 += mn.cnt2
timeAc3 += mn.cnt3
timeAc4 += mn.cnt4
timeAc1 /= 10
timeAc2 /= 10
timeAc3 /= 10
timeAc4 /= 10
yAc1.append(timeAc1)
yAc2.append(timeAc2)
yAc3.append(timeAc3)
yAc4.append(timeAc4)
plt.plot(xAc1, yAc1, color='g', label='AC 1')
plt.plot(xAc2, yAc2, color='b', label='AC 2')
plt.plot(xAc3, yAc3, color='r', label='AC 3')
plt.plot(xAc4, yAc4, color='orange', label='AC 4')
plt.xlabel('Number of Node')
plt.ylabel('Number of Satisfactions')
plt.title('Comparison of Arc Consistency Algorithm')
plt.suptitle('@mashrur')
plt.legend(loc='upper left')
plt.savefig('node_satisfaction_small.png')
plt.show()
def visualize_density(sz):
xAc1 = []
yAc1 = []
xAc2 = []
yAc2 = []
xAc3 = []
yAc3 = []
xAc4 = []
yAc4 = []
node = 100
for density in numpy.arange(0.1, 1.1, 0.3):
xAc1.append(density)
xAc2.append(density)
xAc3.append(density)
xAc4.append(density)
timeAc1 = 0
timeAc2 = 0
timeAc3 = 0
timeAc4 = 0
domainSize = sz
constraints = {}
domain = [[0 for x in range(domainSize)] for y in range(node)] # node x domainSize size list
edge = (density*(node*(node-1)))//2
edge = int(edge)
print('lol', density, edge)
g = nw.gnm_random_graph(node, edge, False)
for i in g.edges:
constraints[i] = random.randint(0, 10)
constraints[(i[1], i[0])] = constraints[i] + lim
for i in range(node):
for j in range(domainSize):
domain[i][j] = randInt(-1100, 1100)
timeAc1 += executeAC1(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
timeAc2 += executeAC2(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
timeAc3 += executeAC3(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
timeAc4 += executeAC4(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
yAc1.append(timeAc1)
yAc2.append(timeAc2)
yAc3.append(timeAc3)
yAc4.append(timeAc4)
plt.plot(xAc1, yAc1, color='g', label='AC 1')
plt.plot(xAc2, yAc2, color='b', label='AC 2')
plt.plot(xAc3, yAc3, color='r', label='AC 3')
plt.plot(xAc4, yAc4, color='orange', label='AC 4')
plt.xlabel('Graph Density')
plt.ylabel('Performance (msec)')
plt.title('Comparison of Arc Consistency Algorithm')
plt.suptitle('@mashrur')
plt.legend(loc='upper left')
plt.savefig('density_large_domain.png')
plt.show()
def visualize_nodes(domainSiz):
xAc1 = []
yAc1 = []
xAc2 = []
yAc2 = []
xAc3 = []
yAc3 = []
xAc4 = []
yAc4 = []
for node in range(1, 101, 20):
print(node)
xAc1.append(node)
xAc2.append(node)
xAc3.append(node)
xAc4.append(node)
timeAc1 = 0
timeAc2 = 0
timeAc3 = 0
timeAc4 = 0
for i in range(10):
csp = generateCSP(node, domainSiz)
g = csp[0]
constraints = csp[1]
domain = csp[2]
edge = csp[3]
domainSize = csp[4]
timeAc1 += executeAC1(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
timeAc2 += executeAC2(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
timeAc3 += executeAC3(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
timeAc4 += executeAC4(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
timeAc1 /= 10
timeAc2 /= 10
timeAc3 /= 10
timeAc4 /= 10
yAc1.append(timeAc1)
yAc2.append(timeAc2)
yAc3.append(timeAc3)
yAc4.append(timeAc4)
print('AC 1')
for i in yAc1:
print(i)
print('AC 2')
for i in yAc2:
print(i)
print('AC 3')
for i in yAc3:
print(i)
print('AC 4')
for i in yAc4:
print(i)
plt.plot(xAc1, yAc1, color='g', label='AC 1')
plt.plot(xAc2, yAc2, color='b', label='AC 2')
plt.plot(xAc3, yAc3, color='r', label='AC 3')
plt.plot(xAc4, yAc4, color='orange', label='AC 4')
plt.xlabel('Number of Node')
plt.ylabel('Performance (msec)')
plt.title('Comparison of Arc Consistency Algorithm')
plt.suptitle('@mashrur')
plt.legend(loc='upper left')
#plt.savefig('node_variable_sized_domain.png')
plt.show()
def visualize_isSatEdge(domainSz):
xAc1 = []
yAc1 = []
xAc2 = []
yAc2 = []
xAc3 = []
yAc3 = []
xAc4 = []
yAc4 = []
xAc1.append(0)
xAc2.append(0)
xAc3.append(0)
xAc4.append(0)
yAc1.append(0)
yAc2.append(0)
yAc3.append(0)
yAc4.append(0)
for edge in range(1, 110, 10):
print(edge)
xAc1.append(edge)
xAc2.append(edge)
xAc3.append(edge)
xAc4.append(edge)
timeAc1 = 0
timeAc2 = 0
timeAc3 = 0
timeAc4 = 0
domainSize = domainSz
node = randInt(edge+1, 2*edge)
for p in range(10):
constraints = {}
domain = [[0 for x in range(domainSize)] for y in range(node)] # node x domainSize size list
g = nw.gnm_random_graph(node, edge, False)
for i in g.edges:
constraints[i] = random.randint(0, 10)
constraints[(i[1], i[0])] = constraints[i] + lim
for i in range(node):
for j in range(domainSize):
domain[i][j] = randInt(1, 100)
mn.executeAC1(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
mn.executeAC2(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
mn.executeAC3(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
mn.executeAC4(node, edge, domainSize, deepcopy(g), deepcopy(constraints), deepcopy(domain))
timeAc1 += mn.cnt1
timeAc2 += mn.cnt2
timeAc3 += mn.cnt3
timeAc4 += mn.cnt4
timeAc1 /= 10
timeAc2 /= 10
timeAc3 /= 10
timeAc4 /= 10
yAc1.append(timeAc1)
yAc2.append(timeAc2)
yAc3.append(timeAc3)
yAc4.append(timeAc4)
plt.plot(xAc1, yAc1, color='g', label='AC 1')
plt.plot(xAc2, yAc2, color='b', label='AC 2')
plt.plot(xAc3, yAc3, color='r', label='AC 3')
plt.plot(xAc4, yAc4, color='orange', label='AC 4')
plt.xlabel('Number of Edge')
plt.ylabel('Number of Satisfactions')
plt.title('Comparison of Arc Consistency Algorithm')
plt.suptitle('@mashrur')
plt.legend(loc='upper left')
# plt.savefig('edge_satisfactions_large.png')
plt.show()
def visualize_domainReduction(domainSiz):
xAc1 = []
yAc1 = []
xAc2 = []
yAc2 = []
xAc3 = []
yAc3 = []
xAc4 = []
yAc4 = []
for node in range(1, 101, 20):
print(node)
xAc1.append(node)
xAc2.append(node)
xAc3.append(node)
xAc4.append(node)
csp = generateCSP(node, domainSiz)
g = csp[0]
constraints = csp[1]
domain = csp[2]
edge = csp[3]
domainSize = csp[4]
initSizeAC1 = node*domainSize
initSizeAC2 = node*domainSize
initSizeAC3 = node*domainSize
initSizeAC4 = node*domainSize
finalSizeAC1 = 0
finalSizeAC2 = 0
finalSizeAC3 = 0
finalSizeAC4 = 0
domain1 = deepcopy(domain)
domain2 = deepcopy(domain)
domain3 = deepcopy(domain)
domain4 = deepcopy(domain)
executeAC1(node, edge, domainSize, deepcopy(g), deepcopy(constraints), domain1)
executeAC2(node, edge, domainSize, deepcopy(g), deepcopy(constraints), domain2)
executeAC3(node, edge, domainSize, deepcopy(g), deepcopy(constraints), domain3)
executeAC4(node, edge, domainSize, deepcopy(g), deepcopy(constraints), domain4)
for p in range(node):
finalSizeAC1 += len(domain1[p])
finalSizeAC2 += len(domain2[p])
finalSizeAC3 += len(domain3[p])
finalSizeAC4 += len(domain4[p])
sizeAc1 = abs((finalSizeAC1-initSizeAC1)/initSizeAC1)
sizeAc2 = abs((finalSizeAC2-initSizeAC2)/initSizeAC2)
sizeAc3 = abs((finalSizeAC3-initSizeAC3)/initSizeAC3)
sizeAc4 = abs((finalSizeAC4-initSizeAC4)/initSizeAC4)
print('lol', sizeAc1, sizeAc2, sizeAc3, sizeAc4)
yAc1.append(sizeAc1)
yAc2.append(sizeAc2)
yAc3.append(sizeAc3)
yAc4.append(sizeAc4)
plt.plot(xAc1, yAc1, color='g', label='AC 1')
plt.plot(xAc2, yAc2, color='b', label='AC 2')
plt.plot(xAc3, yAc3, color='r', label='AC 3')
plt.plot(xAc4, yAc4, color='orange', label='AC 4')
plt.xlabel('Number of Node')
plt.ylabel('Domain Reduction')
plt.title('Comparison of Arc Consistency Algorithm')
plt.suptitle('@mashrur')
plt.legend(loc='upper left')
plt.savefig('node_domain_reduction.png')
plt.show()