def test_check_validity(self):
exception_raised = False
try:
pgh.geohash_approximate_distance('shibu', 'shiba', check_validity=True)
except ValueError:
exception_raised = True
self.assertTrue(exception_raised)
def test_distance(self):
# test the fast geohash distance approximations
self.assertEqual(pgh.geohash_approximate_distance('bcd3u', 'bc83n'), 625441)
self.assertEqual(pgh.geohash_approximate_distance('bcd3uasd', 'bcd3n'), 19545)
self.assertEqual(pgh.geohash_approximate_distance('bcd3u', 'bcd3uasd'), 3803)
self.assertEqual(pgh.geohash_approximate_distance('bcd3ua', 'bcd3uasdub'), 610)
# test the haversine great circle distance calculations
self.assertAlmostEqual(pgh.geohash_haversine_distance('testxyz', 'testwxy'), 6339.483649071294, places=4)
def test_check_validity(self):
exception_raised = False
try:
pgh.geohash_approximate_distance('shibu',
'shiba',
check_validity=True)
except ValueError:
exception_raised = True
self.assertTrue(exception_raised)
def test_distance(self):
# test the fast geohash distance approximations
self.assertEqual(pgh.geohash_approximate_distance('bcd3u', 'bc83n'),
625441)
self.assertEqual(pgh.geohash_approximate_distance('bcd3uasd', 'bcd3n'),
19545)
self.assertEqual(pgh.geohash_approximate_distance('bcd3u', 'bcd3uasd'),
3803)
self.assertEqual(
pgh.geohash_approximate_distance('bcd3ua', 'bcd3uasdub'), 610)
# test the haversine great circle distance calculations
self.assertAlmostEqual(pgh.geohash_haversine_distance(
'testxyz', 'testwxy'),
6339.483649071294,
places=4)
def calculateDistance(route):
distance = 0
for i, j in pairwise(route['gp']):
if i and j:
calc = round(pgh.geohash_approximate_distance(i, j) / 1000)
distance = distance + calc
return distance
Location:
- Categorized Long/Lat
https://datascience.stackexchange.com/questions/13567/ways-to-deal-with-longitude-latitude-feature
https://datascience.stackexchange.com/questions/23651/can-gps-coordinates-latitude-and-longitude-be-used-as-features-in-a-linear-mod
- GeoHash:
http://www.movable-type.co.uk/scripts/geohash.html
'''
#Create Geohash for train
import pygeohash as pgh
Train_start_geohash = [pgh.encode(x,y) for x,y in zip(Trainlat1, Trainlon1)]
Train_end_geohash = [pgh.encode(x,y) for x,y in zip(Trainlat2, Trainlon2)]
Train_geodistance = [pgh.geohash_approximate_distance(x,y) for x,y in zip(Train_start_geohash, Train_end_geohash)]
len(Train_start_geohash) #12905715
len(set(Train_start_geohash)) #9753342
len(Train_end_geohash) #12905715
len(set(Train_end_geohash)) #10832199
len(Train_geodistance) #12905715
len(set(Train_geodistance)) #9
start_geohash_TrainDF = pd.DataFrame({'start_geohash': Train_start_geohash})
end_geohash_TrainDF = pd.DataFrame({'end_geohash': Train_end_geohash})
geodistance_TrainDF = pd.DataFrame({'geoDist': Train_geodistance})
Train = pd.concat([Train, geodistance_TrainDF], axis=1)
def approx_dist(a_lat, a_long, b_lat, b_long):
a = pgh.encode(a_lat, a_long)
b = pgh.encode(b_lat, b_long)
return pgh.geohash_approximate_distance(a, b) / 1000
Location:
- Categorized Long/Lat
- IsBusyLocation(start/end),
https://datascience.stackexchange.com/questions/13567/ways-to-deal-with-longitude-latitude-feature
https://datascience.stackexchange.com/questions/23651/can-gps-coordinates-latitude-and-longitude-be-used-as-features-in-a-linear-mod
- GeoHash:
http://www.movable-type.co.uk/scripts/geohash.html
'''
#Create Geohash for train
import pygeohash as pgh
Train_start_geohash = [pgh.encode(x,y) for x,y in zip(Trainlat1, Trainlon1)]
Train_end_geohash = [pgh.encode(x,y) for x,y in zip(Trainlat2, Trainlon2)]
Train_geodistance = [pgh.geohash_approximate_distance(x,y) for x,y in zip(Train_start_geohash, Train_end_geohash)]
len(Train_start_geohash) #12905715
len(set(Train_start_geohash)) #9753342
len(Train_end_geohash) #12905715
len(set(Train_end_geohash)) #10832199
len(Train_geodistance) #12905715
len(set(Train_geodistance)) #9
start_geohash_TrainDF = pd.DataFrame({'start_geohash': Train_start_geohash})
end_geohash_TrainDF = pd.DataFrame({'end_geohash': Train_end_geohash})
geodistance_TrainDF = pd.DataFrame({'geoDist': Train_geodistance})
Train = pd.concat([Train, geodistance_TrainDF], axis=1)