def HSR_main(date, ref_num=3):
user_data_path = "./data/user_data/"+date+"/" #�s�Ҧ��ϥΪ̪��@�Ѫ���ƪ���Ƨ�
result = []
try:
request_date = date[:4] + '-' + date[4:6] + '-' + date[6:]
except:
print('date format error')
travel_time, travel_during_time = External.HSR_travel_time(request_date)
stations = External.HSR_station()
cell_file = "./data/all_tower.csv" #�s�Ҧ���m���ɮ�(lon,lat)
HSR_ref_sys = Preprocess.HSR_reference_system(cell_file, stations, ref_num)
print("-----build up rail reference system-----")
for file_name in os.listdir(user_data_path):
f = open(os.path.join(user_data_path,file_name),"r")
user_raw_data = [[row[0], row[4], row[2], row[3]] for row in
csv.reader(f, delimiter='|')] # [imsi, unix_time, lon, lat]
raw_data_df = pandas.DataFrame(user_raw_data, columns=['imsi', 'unix_time', 'lon', 'lat'])
raw_data_df = raw_data_df.sort_values(['unix_time'])
user_raw_data_sorted = raw_data_df.values.tolist()
user_data = Preprocess.preprocessing(user_raw_data_sorted) #[imsi, start_time, end_time, lon, lat]
r = Mode_Detection.HSR_trip_detection(user_data, travel_time, travel_during_time, stations, HSR_ref_sys)
if len(r)!=0:
result.extend(r)
output_result = [(result,)]
csv_output(output_result, ['hsr_'+date+'.csv'])
def mrt_main(date):
user_data_path = "./data/user_data/"+date+"/"
rf_sys_path = "./data/external_data/mrt/reference system/entrance_tower_to_subway_k_1.csv"
mrt_route_file = "./data/external_data/mrt/mrt_route.csv"
mrt_entrance_file = "./data/external_data/mrt/mrt_station_entrance.csv"
travel_time_file = "./data/external_data/mrt/mrt_travel_time.csv"
result = []
for file_name in os.listdir(user_data_path):
#new part
f = open(os.path.join(user_data_path,file_name),"r")
user_raw_data = [[row[0], row[4], row[2], row[3]] for row in
csv.reader(f, delimiter='|')] # [imsi, unix_time, lon, lat]
raw_data_df = pandas.DataFrame(user_raw_data, columns=['imsi', 'unix_time', 'lon', 'lat'])
raw_data_df = raw_data_df.sort_values(['unix_time'])
user_raw_data_sorted = raw_data_df.values.tolist()
f.close()
user_data = Preprocess.preprocessing(user_raw_data_sorted) #[imsi, start_time, end_time, lon, lat]
# print(user_data)
r = Mode_Detection.MRT_trip_detection(user_data,rf_sys_path,mrt_route_file,mrt_entrance_file,travel_time_file)
if len(r) == 0:
continue
if len(r[0]) == 0:
continue
result.append(r)
csv_output(result, ['mrt_'+date+'.csv', 'mrt_path_'+date+'.csv'])
def bus_main(date):
user_data_path = "./data/user_data/"+date+"/"
route = pickle.load(open("./data/external_data/bus/bus_route.pickle", "rb"))
all_user_data = dict()
for file_name in os.listdir(user_data_path):
f = open(os.path.join(user_data_path, file_name),"r")
user_raw_data = [[row[0], row[4], row[2], row[3]] for row in csv.reader(f, delimiter='|')] #[imsi, unix_time, lon, lat]
raw_data_df = pandas.DataFrame(user_raw_data, columns=['imsi', 'time', 'lon', 'lat'])
raw_data_df = raw_data_df.sort_values(['time'])
user_raw_data_sorted = raw_data_df.values.tolist()
user_data = Preprocess.preprocessing(user_raw_data_sorted) #[imsi, start_time, end_time, lon, lat]
all_user_data[user_data[0][0]] = user_data
f.close()
'''
TPE_route = External.bus_route('taipei')
NWT_route = External.bus_route('newtaipei')
route = TPE_route
for key in NWT_route.keys():
route[key] = NWT_route[key]
pickle.dump(route, open("data/bus_route.pickle", "wb"))
'''
rid2user,user2rid,route2rid = Preprocess.bus_spatial_index(route, all_user_data)
SpeedDis = [20]*24
result = Mode_Detection.bus_trip_detection(rid2user, user2rid, route2rid, route, SpeedDis)
#print result
output_result = [(result,)]
csv_output(output_result, ['bus_'+date+'.csv'])
def rail_main(date):
user_data_path = "./data/user_data/"+date+"/"
result = []
# travel_time, travel_during_time = External.rail_travel_time(date)
stations = External.rail_station()
cell_file = "./data/all_tower.csv" #format (lon,lat)
rail_ref_sys = Preprocess.rail_reference_system(cell_file, stations)
#print("-----build up rail reference system-----")
for file_name in os.listdir(user_data_path):
f = open(os.path.join(user_data_path,file_name),"r")
user_raw_data = [[row[0], row[4], row[2], row[3]] for row in
csv.reader(f, delimiter='|')] # [imsi, unix_time, lon, lat]
raw_data_df = pandas.DataFrame(user_raw_data, columns=['imsi', 'unix_time', 'lon', 'lat'])
raw_data_df = raw_data_df.sort_values(['unix_time'])
user_raw_data_sorted = raw_data_df.values.tolist()
user_data = Preprocess.preprocessing(user_raw_data_sorted) #[imsi, start_time, end_time, lon, lat]
r = Mode_Detection.rail_trip_detection(user_data, stations, rail_ref_sys)
if len(r)!=0:
result.extend(r)
output_result = [(result,)]
csv_output(output_result, ['train_'+date+'.csv'])