def instance_avgsk():
print('=== Test how instance count affect candidate skyline ===')
inst = [3, 4, 5, 6, 7, 8, 9, 10]
for ins in inst:
dqueue = batchImport('10000_dim2_pos' + str(ins) + '_rad5_01000.csv',
ins)
print('========== instance count = ' + str(ins) + ' ==========')
print('---------- Brute force ----------')
tbsky = slideBPSky(2, ins, 5, [0, 1000], wsize=300)
avgsk1, avgsk2 = 0, 0
for i in range(10000):
tbsky.receiveData(dqueue[i])
tbsky.updateSkyline()
avgsk1 += len(tbsky.getSkyline())
avgsk2 += len(tbsky.getSkyline2())
tbsky.removeRtree()
avgsk1, avgsk2 = avgsk1 / 10000, avgsk2 / 10000
print('Avg. sky1: ' + str(avgsk1))
print('Avg. sky2: ' + str(avgsk2))
print('---------- Update ----------')
tusky = slideUPSky(2, ins, 5, [0, 1000], wsize=300)
avgsk1, avgsk2 = 0, 0
for i in range(10000):
tusky.receiveData(dqueue[i])
tusky.updateSkyline()
avgsk1 += len(tusky.getSkyline())
avgsk2 += len(tusky.getSkyline2())
tusky.removeRtree()
avgsk1, avgsk2 = avgsk1 / 10000, avgsk2 / 10000
print('Avg. sky1: ' + str(avgsk1))
print('Avg. sky2: ' + str(avgsk2))
def radius_avgsk():
print('=== Test how data radius affect candidate skyline ===')
radius = [3, 4, 5, 6, 7, 8, 9, 10]
for r in radius:
dqueue = batchImport('10000_dim2_pos5_rad' + str(r) + '_01000.csv', 5)
print('========== radius = ' + str(r) + ' ==========')
print('---------- Brute force ----------')
tbsky = slideBPSky(2, 5, r, [0, 1000], wsize=300)
avgsk1, avgsk2 = 0, 0
for i in range(10000):
tbsky.receiveData(dqueue[i])
tbsky.updateSkyline()
avgsk1 += len(tbsky.getSkyline())
avgsk2 += len(tbsky.getSkyline2())
tbsky.removeRtree()
avgsk1, avgsk2 = avgsk1 / 10000, avgsk2 / 10000
print('Avg. sky1: ' + str(avgsk1))
print('Avg. sky2: ' + str(avgsk2))
print('---------- Update ----------')
tusky = slideUPSky(2, 5, r, [0, 1000], wsize=300)
avgsk1, avgsk2 = 0, 0
for i in range(10000):
tusky.receiveData(dqueue[i])
tusky.updateSkyline()
avgsk1 += len(tusky.getSkyline())
avgsk2 += len(tusky.getSkyline2())
tusky.removeRtree()
avgsk1, avgsk2 = avgsk1 / 10000, avgsk2 / 10000
print('Avg. sky1: ' + str(avgsk1))
print('Avg. sky2: ' + str(avgsk2))
def wsize_avgsk():
print("=== Test how window size affect candidate skyline ===")
wsize = [100, 200, 300, 400, 500, 600, 700, 800, 900, 1000]
dqueue = batchImport('10000_dim2_pos5_rad5_01000.csv', 5)
for w in wsize:
print('========== window size = ' + str(w) + ' ==========')
print('---------- Brute force ----------')
tbsky = slideBPSky(2, 5, 5, [0, 1000], wsize=w)
avgsk1, avgsk2 = 0, 0
for i in range(10000):
tbsky.receiveData(dqueue[i])
tbsky.updateSkyline()
avgsk1 += len(tbsky.getSkyline())
avgsk2 += len(tbsky.getSkyline2())
tbsky.removeRtree()
avgsk1, avgsk2 = avgsk1 / 10000, avgsk2 / 10000
print('Avg. sky1: ' + str(avgsk1))
print('Avg. sky2: ' + str(avgsk2))
print('---------- Update ----------')
tusky = slideUPSky(2, 5, 5, [0, 1000], wsize=w)
avgsk1, avgsk2 = 0, 0
for i in range(10000):
tusky.receiveData(dqueue[i])
tusky.updateSkyline()
avgsk1 += len(tusky.getSkyline())
avgsk2 += len(tusky.getSkyline2())
tusky.removeRtree()
avgsk1, avgsk2 = avgsk1 / 10000, avgsk2 / 10000
print('Avg. sky1: ' + str(avgsk1))
print('Avg. sky2: ' + str(avgsk2))
def dim_avgsk():
print("=== Test how dimension of data affect candidate size ===")
dim = [2, 3, 4, 5, 6, 7, 8, 9, 10]
for d in dim:
dqueue = batchImport('10000_dim'+str(d)+'_pos5_rad5_01000.csv', 5)
print('========== Data dimension = '+ str(d) + ' ==========')
print('---------- Brute force ----------')
tbsky = slideBPSky(d, 5, 5, [0,1000], wsize=300)
avgsk1, avgsk2 = 0, 0
for i in range(10000):
tbsky.receiveData(dqueue[i])
tbsky.updateSkyline()
avgsk1 += len(tbsky.getSkyline())
avgsk2 += len(tbsky.getSkyline2())
tbsky.removeRtree()
avgsk1, avgsk2 = avgsk1/10000, avgsk2/10000
print('Avg. sky1: '+ str(avgsk1))
print('Avg. sky2: '+ str(avgsk2))
print('---------- Update ----------')
tusky = slideUPSky(d, 5, 5, [0,1000], wsize=300)
avgsk1, avgsk2 = 0, 0
for i in range(10000):
tusky.receiveData(dqueue[i])
tusky.updateSkyline()
avgsk1 += len(tusky.getSkyline())
avgsk2 += len(tusky.getSkyline2())
tusky.removeRtree()
avgsk1, avgsk2 = avgsk1/10000, avgsk2/10000
print('Avg. sky1: '+ str(avgsk1))
print('Avg. sky2: '+ str(avgsk2))
def radius_time():
print('=== Test how data radius affect running time ===')
radius = [3, 4, 5, 6, 7, 8, 9, 10]
for r in radius:
dqueue = batchImport('10000_dim2_pos5_rad' + str(r) + '_01000.csv', 5)
print('========== radius = ' + str(r) + ' ==========\n')
print('---------- Brute force ----------')
tbsky = slideBPSky(2, 5, r, [0, 1000], wsize=300)
start_time = time.time()
for i in range(10000):
tbsky.receiveData(dqueue[i])
tbsky.updateSkyline()
tbsky.removeRtree()
print("--- %s seconds ---" % (time.time() - start_time))
print('---------- Update ----------')
tusky = slideUPSky(2, 5, r, [0, 1000], wsize=300)
start_time = time.time()
for i in range(10000):
tusky.receiveData(dqueue[i])
tusky.updateSkyline()
tusky.removeRtree()
print("--- %s seconds ---" % (time.time() - start_time))
def wsize_time():
print("=== Test how window size affect running time ===")
wsize = [100, 200, 300, 400, 500, 600, 700, 800, 900, 1000]
dqueue = batchImport('10000_dim2_pos5_rad5_01000.csv', 5)
for w in wsize:
print('========== window size = ' + str(w) + ' ==========')
print('---------- Brute force ----------')
tbsky = slideBPSky(2, 5, 5, [0, 1000], wsize=w)
start_time = time.time()
for i in range(10000):
tbsky.receiveData(dqueue[i])
tbsky.updateSkyline()
tbsky.removeRtree()
print("--- %s seconds ---" % (time.time() - start_time))
print('---------- Update ----------')
tusky = slideUPSky(2, 5, 5, [0, 1000], wsize=w)
start_time = time.time()
for i in range(10000):
tusky.receiveData(dqueue[i])
tusky.updateSkyline()
tusky.removeRtree()
print("--- %s seconds ---" % (time.time() - start_time))
def dim_time():
print("=== Test how dimension of data affect running time ===")
dim = [2, 3, 4, 5, 6, 7, 8, 9, 10]
for d in dim:
dqueue = batchImport('10000_dim'+str(d)+'_pos5_rad5_01000.csv', 5)
print('========== Data dimension = '+ str(d) + ' ==========')
print('---------- Brute force ----------')
tbsky = slideBPSky(d, 5, 5, [0,1000], wsize=300)
start_time = time.time()
for i in range(10000):
tbsky.receiveData(dqueue[i])
tbsky.updateSkyline()
tbsky.removeRtree()
print("--- %s seconds ---" % (time.time() - start_time))
print('---------- Update ----------')
tusky = slideUPSky(d, 5, 5, [0,1000], wsize=300)
start_time = time.time()
for i in range(10000):
tusky.receiveData(dqueue[i])
tusky.updateSkyline()
tusky.removeRtree()
print("--- %s seconds ---" % (time.time() - start_time))
def instance_time():
print('=== Test how instance count affect running time ===')
inst = [3, 4, 5, 6, 7, 8, 9, 10]
for ins in inst:
dqueue = batchImport('10000_dim2_pos' + str(ins) + '_rad5_01000.csv',
ins)
print('========== instance count = ' + str(ins) + ' ==========')
print('---------- Brute force ----------')
tbsky = slideBPSky(2, ins, 5, [0, 1000], wsize=300)
start_time = time.time()
for i in range(10000):
tbsky.receiveData(dqueue[i])
tbsky.updateSkyline()
tbsky.removeRtree()
print("--- %s seconds ---" % (time.time() - start_time))
print('---------- Update ----------')
tusky = slideUPSky(2, ins, 5, [0, 1000], wsize=300)
start_time = time.time()
for i in range(10000):
tusky.receiveData(dqueue[i])
tusky.updateSkyline()
tusky.removeRtree()
print("--- %s seconds ---" % (time.time() - start_time))