def testBinData(self):
sim = similarity.similarity([], 300)
self.assertTrue(not sim.bins)
sim = similarity.similarity(self.data, 300)
sim.bin_data()
sum = 0
for bin in sim.bins:
sum += len(bin)
self.assertTrue(sum == len(sim.data))
testbins = set()
for bin in sim.bins:
for b in bin:
testbins.add(b)
self.assertTrue(len(testbins) == len(sim.data))
for i in range(len(sim.bins) - 1):
self.assertTrue(len(sim.bins[i]) >= len(sim.bins[i + 1]))
data = []
now = time.time()
start = [-122, 47]
end = [-123, 47]
for i in range(10):
data.append(etatc._createTripEntry(self, now, now, start, end))
start = [-74, 41]
end = [-74, 42]
for i in range(10):
data.append(etatc._createTripEntry(self, now, now, start, end))
sim = similarity.similarity(data, 300)
sim.bin_data()
self.assertTrue(len(sim.bins) == 2)
def testBinData(self):
sim = similarity.similarity([], 300)
self.assertTrue(not sim.bins)
sim = similarity.similarity(self.data, 300)
sim.bin_data()
sum = 0
for bin in sim.bins:
sum += len(bin)
self.assertTrue(sum == len(sim.data))
testbins = set()
for bin in sim.bins:
for b in bin:
testbins.add(b)
self.assertTrue(len(testbins) == len(sim.data))
for i in range(len(sim.bins)-1):
self.assertTrue(len(sim.bins[i]) >= len(sim.bins[i+1]))
data = []
now = time.time()
start = [-122, 47]
end = [-123, 47]
for i in range(10):
data.append(etatc._createTripEntry(self, now, now, start, end))
start = [-74, 41]
end = [-74, 42]
for i in range(10):
data.append(etatc._createTripEntry(self, now, now, start, end))
sim = similarity.similarity(data, 300)
sim.bin_data()
self.assertTrue(len(sim.bins) == 2)
def testDistance(self):
start = [-122.259447, 37.875174]
end1 = [-122.259279, 37.875479]
end2 = [-122.252287, 37.869569]
now = time.time()
t1 = etatc._createTripEntry(self, now, now, start, end1)
t2 = etatc._createTripEntry(self, now, now, start, end2)
sim = similarity.similarity(self.data, 300)
self.assertTrue(sim.distance(start[1], start[0], end1[1], end1[0]))
self.assertTrue(not sim.distance(start[1], start[0], end2[1], end2[0]))
def testDistance(self):
start = [-122.259447, 37.875174]
end1 = [-122.259279, 37.875479]
end2 = [-122.252287, 37.869569]
now = time.time()
t1 = etatc._createTripEntry(self, now, now, start, end1)
t2 = etatc._createTripEntry(self, now, now, start, end2)
sim = similarity.similarity(self.data, 300)
self.assertTrue(sim.distance(start[1], start[0], end1[1], end1[0]))
self.assertTrue(not sim.distance(start[1], start[0], end2[1], end2[0]))
def testLocations(self):
repy = rep.representatives(self.data, self.labels)
repy.list_clusters()
repy.get_reps()
repy.locations()
total = 0
for bin in repy.bins:
for i in range(len(bin)):
b = bin[i]
if b[0] == 'start':
a = repy.reps[b[1]].data.start_loc
if b[0] == 'end':
a = repy.reps[b[1]].data.end_loc
for j in range(i):
c = bin[j]
if c[0] == 'start':
d = repy.reps[c[1]].data.start_loc
if c[0] == 'end':
d = repy.reps[c[1]].data.end_loc
self.assertTrue(
repy.distance(a.coordinates[1], a.coordinates[0], d.
coordinates[1], d.coordinates[0]) < 300)
total += len(bin)
self.assertTrue(total == 2 * repy.num_clusters)
for i in range(repy.num_clusters):
self.assertTrue(
sum(bin.count(('start', i)) for bin in repy.bins) == 1)
self.assertTrue(
sum(bin.count(('end', i)) for bin in repy.bins) == 1)
self.assertTrue(len(repy.locs) == len(repy.bins))
now = time.time()
tripa = etatc._createTripEntry(self, now, now, [1, 2], [30, 40])
tripb = etatc._createTripEntry(self, now, now, [1.0000002, 2.0000002],
[55.0000002, 85.0000002])
tripc = etatc._createTripEntry(self, now, now,
[30.0000002, 40.0000002], [55, 85])
data = [tripa, tripb, tripc]
labels = [0, 1, 2]
repy = rep.representatives(data, labels)
repy.list_clusters()
repy.get_reps()
repy.locations()
self.assertEqual(repy.bins[0], [('start', 0), ('start', 1)])
self.assertEqual(repy.bins[1], [('end', 0), ('start', 2)])
self.assertEqual(repy.bins[2], [('end', 1), ('end', 2)])
self.assertAlmostEqual(repy.locs[0][0], 1.0000001, places=7)
self.assertAlmostEqual(repy.locs[0][1], 2.0000001, places=7)
self.assertAlmostEqual(repy.locs[1][0], 30.0000001, places=7)
self.assertAlmostEqual(repy.locs[1][1], 40.0000001, places=7)
self.assertAlmostEqual(repy.locs[2][0], 55.0000001, places=7)
self.assertAlmostEqual(repy.locs[2][1], 85.0000001, places=7)
def testLocations(self):
repy = rep.representatives(self.data, self.labels)
repy.list_clusters()
repy.get_reps()
repy.locations()
total = 0
for bin in repy.bins:
for i in range(len(bin)):
b = bin[i]
if b[0] == 'start':
a = repy.reps[b[1]].data.start_loc
if b[0] == 'end':
a = repy.reps[b[1]].data.end_loc
for j in range(i):
c = bin[j]
if c[0] == 'start':
d = repy.reps[c[1]].data.start_loc
if c[0] == 'end':
d = repy.reps[c[1]].data.end_loc
self.assertTrue(repy.distance(a.coordinates[1], a.coordinates[0],
d.coordinates[1], d.coordinates[0]) < 300)
total += len(bin)
self.assertTrue(total == 2 * repy.num_clusters)
for i in range(repy.num_clusters):
self.assertTrue(sum(bin.count(('start',i)) for bin in repy.bins) == 1)
self.assertTrue(sum(bin.count(('end',i)) for bin in repy.bins) == 1)
self.assertTrue(len(repy.locs) == len(repy.bins))
now = time.time()
tripa = etatc._createTripEntry(self, now, now, [1,2], [30,40])
tripb = etatc._createTripEntry(self, now, now, [1.0000002,2.0000002], [55.0000002,85.0000002])
tripc = etatc._createTripEntry(self, now, now, [30.0000002,40.0000002], [55,85])
data = [tripa, tripb, tripc]
labels = [0,1,2]
repy = rep.representatives(data, labels)
repy.list_clusters()
repy.get_reps()
repy.locations()
self.assertEqual(repy.bins[0], [('start', 0), ('start', 1)])
self.assertEqual(repy.bins[1], [('end', 0), ('start', 2)])
self.assertEqual(repy.bins[2], [('end', 1), ('end', 2)])
self.assertAlmostEqual(repy.locs[0][0], 1.0000001, places=7)
self.assertAlmostEqual(repy.locs[0][1], 2.0000001, places=7)
self.assertAlmostEqual(repy.locs[1][0], 30.0000001, places=7)
self.assertAlmostEqual(repy.locs[1][1], 40.0000001, places=7)
self.assertAlmostEqual(repy.locs[2][0], 55.0000001, places=7)
self.assertAlmostEqual(repy.locs[2][1], 85.0000001, places=7)
def testMatch(self):
now = time.time()
tripa = etatc._createTripEntry(self, now, now, [1,2], [3,4])
tripb = etatc._createTripEntry(self, now, now, [3,4], [1,2])
tripc = etatc._createTripEntry(self, now, now, [1,2], [9,10])
data = [tripa, tripb, tripc]
labels = [0,1,2]
repy = rep.representatives(data, labels)
repy.list_clusters()
repy.get_reps()
bin = [('start', 0), ('start', 2)]
self.assertTrue(repy.match('end', 1, bin))
bin = [('start',0), ('end', 0)]
self.assertTrue(not repy.match('end', 1, bin))
bin = [('start', 0), ('start', 1)]
self.assertTrue(not repy.match('end', 2, bin))
def testMatch(self):
now = time.time()
tripa = etatc._createTripEntry(self, now, now, [1, 2], [3, 4])
tripb = etatc._createTripEntry(self, now, now, [3, 4], [1, 2])
tripc = etatc._createTripEntry(self, now, now, [1, 2], [9, 10])
data = [tripa, tripb, tripc]
labels = [0, 1, 2]
repy = rep.representatives(data, labels)
repy.list_clusters()
repy.get_reps()
bin = [('start', 0), ('start', 2)]
self.assertTrue(repy.match('end', 1, bin))
bin = [('start', 0), ('end', 0)]
self.assertTrue(not repy.match('end', 1, bin))
bin = [('start', 0), ('start', 1)]
self.assertTrue(not repy.match('end', 2, bin))
def testElbowDistance(self):
start = [-122, 47]
end = [-123, 47]
now = time.time()
t = etatc._createTripEntry(self, now, now, start, end)
data = [t] * 11
bins = [[1, 2, 3, 4], [5, 6, 10], [7], [8], [9], [0]]
sim = similarity.similarity(data, 300)
sim.bins = bins
sim.delete_bins()
self.assertTrue(sim.num == 2)
def testElbowDistance(self):
start = [-122,47]
end = [-123,47]
now = time.time()
t = etatc._createTripEntry(self, now, now, start, end)
data = [t] * 11
bins = [[1,2,3,4], [5,6,10], [7], [8], [9], [0]]
sim = similarity.similarity(data, 300)
sim.bins = bins
sim.delete_bins()
self.assertTrue(sim.num == 2)
def testReps(self):
repy = rep.representatives(self.data, self.labels)
repy.list_clusters()
repy.get_reps()
self.assertTrue(len(repy.reps) == len(repy.clusters))
clusters = [0]
now = time.time()
tripa = etatc._createTripEntry(self, now, now, [1,2], [3,4])
tripb = etatc._createTripEntry(self, now, now, [9,10], [5,8])
tripc = etatc._createTripEntry(self, now, now, [5,6], [4,6])
data = [tripa, tripb, tripc]
labels = [0,0,0]
repy = rep.representatives(data, labels)
repy.list_clusters()
repy.get_reps()
logging.debug("repy.reps[0].data.start_loc = %s" % repy.reps[0].data.start_loc)
self.assertEqual(repy.reps[0].data.start_loc.coordinates[0], 5)
self.assertEqual(repy.reps[0].data.start_loc.coordinates[1], 6)
self.assertEqual(repy.reps[0].data.end_loc.coordinates[0], 4)
self.assertEqual(repy.reps[0].data.end_loc.coordinates[1], 6)
def testInit(self):
try:
sim = similarity.similarity([], 'b')
except ValueError:
self.assertTrue(True)
except Exception:
self.assertTrue(False)
logging.debug("STARTING init test")
sim = similarity.similarity([], 100)
self.assertTrue(len(sim.data) == 0)
now = time.time()
start = [-122, 47]
end = [-123, 47]
t1 = etatc._createTripEntry(self, now, now, start, start)
t2 = etatc._createTripEntry(self, now, now, start, end)
sim = similarity.similarity([t1, t2], 100)
logging.debug("sim.data = %s" % sim.data)
simmy = similarity.similarity([t2], 100)
logging.debug("simmy.data = %s" % simmy.data)
self.assertTrue(sim.data == simmy.data)
def testReps(self):
repy = rep.representatives(self.data, self.labels)
repy.list_clusters()
repy.get_reps()
self.assertTrue(len(repy.reps) == len(repy.clusters))
clusters = [0]
now = time.time()
tripa = etatc._createTripEntry(self, now, now, [1, 2], [3, 4])
tripb = etatc._createTripEntry(self, now, now, [9, 10], [5, 8])
tripc = etatc._createTripEntry(self, now, now, [5, 6], [4, 6])
data = [tripa, tripb, tripc]
labels = [0, 0, 0]
repy = rep.representatives(data, labels)
repy.list_clusters()
repy.get_reps()
logging.debug("repy.reps[0].data.start_loc = %s" %
repy.reps[0].data.start_loc)
self.assertEqual(repy.reps[0].data.start_loc.coordinates[0], 5)
self.assertEqual(repy.reps[0].data.start_loc.coordinates[1], 6)
self.assertEqual(repy.reps[0].data.end_loc.coordinates[0], 4)
self.assertEqual(repy.reps[0].data.end_loc.coordinates[1], 6)
def testInit(self):
try:
sim = similarity.similarity([], 'b')
except ValueError:
self.assertTrue(True)
except Exception:
self.assertTrue(False)
logging.debug("STARTING init test")
sim = similarity.similarity([], 100)
self.assertTrue(len(sim.data) == 0)
now = time.time()
start = [-122,47]
end = [-123,47]
t1 = etatc._createTripEntry(self, now, now, start, start)
t2 = etatc._createTripEntry(self, now, now, start, end)
sim = similarity.similarity([t1, t2], 100)
logging.debug("sim.data = %s" % sim.data)
simmy = similarity.similarity([t2], 100)
logging.debug("simmy.data = %s" % simmy.data)
self.assertTrue(sim.data == simmy.data)
def testCluster(self):
feat = featurization.featurization(self.data)
feat.cluster(min_clusters=2, max_clusters=10)
self.assertTrue(len(feat.labels) == len(feat.points))
self.assertTrue(feat.clusters == len(set(feat.labels)))
a = feat.cluster(name='kmeans', min_clusters=5, max_clusters=20)
self.assertTrue(len(feat.labels) == len(feat.points))
self.assertTrue(feat.clusters == len(set(feat.labels)))
b = feat.cluster(name='nonname', min_clusters=5, max_clusters=20)
self.assertTrue(
a == b) #defaults to kmeans with invalid clustering method
feat.cluster(min_clusters=len(self.data) + 1)
c = feat.cluster(min_clusters=0, max_clusters=20)
d = feat.cluster(min_clusters=2, max_clusters=20)
self.assertTrue(c == d)
try:
feat.cluster(min_clusters=10, max_clusters=2)
except ValueError:
self.assertTrue(True)
except Exception:
self.assertTrue(False)
data = []
start = [-122, 47]
end = [-123, 47]
now = time.time()
for i in range(10):
a = etatc._createTripEntry(self, now, now, start, end)
data.append(a)
start = [-74, 41]
end = [-74, 42]
for i in range(10):
a = etatc._createTripEntry(self, now, now, start, end)
data.append(a)
feat = featurization.featurization(data)
feat.cluster()
self.assertTrue(len(set(feat.labels)) == 2)
def testCluster(self):
feat = featurization.featurization(self.data)
feat.cluster(min_clusters=2, max_clusters=10)
self.assertTrue(len(feat.labels) == len(feat.points))
self.assertTrue(feat.clusters == len(set(feat.labels)))
a = feat.cluster(name='kmeans', min_clusters=5, max_clusters=20)
self.assertTrue(len(feat.labels) == len(feat.points))
self.assertTrue(feat.clusters == len(set(feat.labels)))
b = feat.cluster(name='nonname', min_clusters=5, max_clusters=20)
self.assertTrue(a == b) #defaults to kmeans with invalid clustering method
feat.cluster(min_clusters=len(self.data)+1)
c = feat.cluster(min_clusters = 0, max_clusters=20)
d = feat.cluster(min_clusters = 2, max_clusters=20)
self.assertTrue(c == d)
try:
feat.cluster(min_clusters = 10, max_clusters=2)
except ValueError:
self.assertTrue(True)
except Exception:
self.assertTrue(False)
data = []
start = [-122, 47]
end = [-123,47]
now = time.time()
for i in range(10):
a = etatc._createTripEntry(self, now, now, start, end)
data.append(a)
start = [-74, 41]
end = [-74, 42]
for i in range(10):
a = etatc._createTripEntry(self, now, now, start, end)
data.append(a)
feat = featurization.featurization(data)
feat.cluster()
self.assertTrue(len(set(feat.labels)) == 2)
def testCalculatePoints(self):
feat = featurization.featurization([])
self.assertTrue(not feat.data)
feat = featurization.featurization(None)
self.assertTrue(not feat.data)
trip = etatc._createTripEntry(self, None, None, None, None)
data = [trip]
try:
feat = featurization.featurization(data)
except AttributeError:
self.assertTrue(True)
except Exception:
self.assertTrue(False)
feat = featurization.featurization(self.data)
self.assertTrue(len(feat.points) == len(feat.data))
for p in feat.points:
self.assertTrue(None not in p)
def testCalculatePoints(self):
feat = featurization.featurization([])
self.assertTrue(not feat.data)
feat = featurization.featurization(None)
self.assertTrue(not feat.data)
trip = etatc._createTripEntry(self, None, None, None, None)
data = [trip]
try:
feat = featurization.featurization(data)
except AttributeError:
self.assertTrue(True)
except Exception:
self.assertTrue(False)
feat = featurization.featurization(self.data)
self.assertTrue(len(feat.points) == len(feat.data))
for p in feat.points:
self.assertTrue(None not in p)