def run():
check_dir_create(summary_dir)
#
Y09 = pd.read_csv("%s/%s" % (airport_trips_dir, "Y09-airport-trip.csv"))
Y10 = pd.read_csv("%s/%s" % (airport_trips_dir, "Y10-airport-trip.csv"))
Y09 = Y09[(Y09["did"] != -1)]
Y10 = Y10[(Y10["did"] != -1)]
#
Y09_mm_did_gb, Y10_mm_did_gb = Y09.groupby(["mm", "did"]), Y10.groupby(["mm", "did"])
Y09_ap_fares = [x / CENT for x in list(Y09_mm_did_gb.sum()["fare"])]
Y10_ap_fares = [x / CENT for x in list(Y10_mm_did_gb.sum()["fare"])]
#
save_pickle_file(driver_monthly_fare_ap_trips, [Y09_ap_fares, Y10_ap_fares])
def run():
check_dir_create(summary_dir)
#
Y09 = pd.read_csv('%s/%s' % (airport_trips_dir, 'Y09-airport-trip.csv'))
Y10 = pd.read_csv('%s/%s' % (airport_trips_dir, 'Y10-airport-trip.csv'))
Y09 = Y09[(Y09['did'] != -1)]
Y10 = Y10[(Y10['did'] != -1)]
#
Y09_mm_did_gb, Y10_mm_did_gb = Y09.groupby(['mm', 'did'
]), Y10.groupby(['mm', 'did'])
Y09_ap_fares = [x / CENT for x in list(Y09_mm_did_gb.sum()['fare'])]
Y10_ap_fares = [x / CENT for x in list(Y10_mm_did_gb.sum()['fare'])]
#
save_pickle_file(driver_monthly_fare_ap_trips,
[Y09_ap_fares, Y10_ap_fares])
def run():
check_dir_create(summary_dir)
#
Y09_driver_total_monthly_fare, Y10_driver_total_monthly_fare = [], []
for y in xrange(9, 11):
for m in xrange(1, 13):
yymm = "%02d%02d" % (y, m)
if yymm in ["0912", "1010"]:
continue
trip_df = pd.read_csv("%s/whole-trip-%s.csv" % (trips_dir, yymm))
trip_df = trip_df[(trip_df["did"] != -1)]
fares = [x / CENT for x in list(trip_df.groupby(["did"]).sum()["fare"])]
if y == 9:
Y09_driver_total_monthly_fare += fares
else:
Y10_driver_total_monthly_fare += fares
save_pickle_file(monthly_fare_summary, [Y09_driver_total_monthly_fare, Y10_driver_total_monthly_fare])
def run():
check_dir_create(tables_dir)
write_text_file(ftd_overall_analysis, 'Init', True)
#
# Load data
#
y09_gen_m_prod, y10_gen_m_prod = load_picle_file(ftd_general_prod_mb)
(y09_piap_m_prod, y09_piap_m_eco_prof), \
(y10_piap_m_prod, y10_piap_m_eco_prof), \
(y09_poap_m_prod, y09_poap_m_eco_prof), \
(y10_poap_m_prod, y10_poap_m_eco_prof) = load_picle_file(ftd_ap_prod_eco_prof_mb)
(y09_pins_m_prod, y09_pins_m_eco_prof), \
(y10_pins_m_prod, y10_pins_m_eco_prof), \
(y09_pons_m_prod, y09_pons_m_eco_prof), \
(y10_pons_m_prod, y10_pons_m_eco_prof) = load_picle_file(ftd_ns_prod_eco_prof_mb)
#
my_d = {
# (Measure, Year): data
('General prod.', 'Y09'): y09_gen_m_prod,
('General prod.', 'Y10'): y10_gen_m_prod,
#
('AP in prod.', 'Y09'): y09_piap_m_prod,
('AP in prod.', 'Y10'): y10_piap_m_prod,
('AP out prod.', 'Y09'): y09_poap_m_prod,
('AP out prod.', 'Y10'): y10_poap_m_prod,
('AP in eco.', 'Y09'): y09_piap_m_eco_prof,
('AP in eco.', 'Y10'): y10_piap_m_eco_prof,
('AP out eco.', 'Y09'): y09_poap_m_eco_prof,
('AP out eco.', 'Y10'): y10_poap_m_eco_prof,
#
('NS in prod.', 'Y09'): y09_pins_m_prod,
('NS in prod.', 'Y10'): y10_pins_m_prod,
('NS out prod.', 'Y09'): y09_pons_m_prod,
('NS out prod.', 'Y10'): y10_pons_m_prod,
('NS in eco.', 'Y09'): y09_pins_m_eco_prof,
('NS in eco.', 'Y10'): y10_pins_m_eco_prof,
('NS out eco.', 'Y09'): y09_pons_m_eco_prof,
('NS out eco.', 'Y10'): y10_pons_m_eco_prof,
}
#
# both_year_comparison(my_d)
draw_month_change(my_d)
def run():
check_dir_create(tables_dir)
write_text_file(ftd_overall_analysis, 'Init', True)
#
# Load data
#
y09_gen_m_prod, y10_gen_m_prod = load_picle_file(ftd_general_prod_mb)
(y09_piap_m_prod, y09_piap_m_eco_prof), \
(y10_piap_m_prod, y10_piap_m_eco_prof), \
(y09_poap_m_prod, y09_poap_m_eco_prof), \
(y10_poap_m_prod, y10_poap_m_eco_prof) = load_picle_file(ftd_ap_prod_eco_prof_mb)
(y09_pins_m_prod, y09_pins_m_eco_prof), \
(y10_pins_m_prod, y10_pins_m_eco_prof), \
(y09_pons_m_prod, y09_pons_m_eco_prof), \
(y10_pons_m_prod, y10_pons_m_eco_prof) = load_picle_file(ftd_ns_prod_eco_prof_mb)
#
my_d = {
# (Measure, Year): data
('General prod.', 'Y09'): y09_gen_m_prod,
('General prod.', 'Y10'): y10_gen_m_prod,
#
('AP in prod.' , 'Y09'): y09_piap_m_prod,
('AP in prod.' , 'Y10'): y10_piap_m_prod,
('AP out prod.' , 'Y09'): y09_poap_m_prod,
('AP out prod.' , 'Y10'): y10_poap_m_prod,
('AP in eco.' , 'Y09'): y09_piap_m_eco_prof,
('AP in eco.' , 'Y10'): y10_piap_m_eco_prof,
('AP out eco.' , 'Y09'): y09_poap_m_eco_prof,
('AP out eco.' , 'Y10'): y10_poap_m_eco_prof,
#
('NS in prod.' , 'Y09'): y09_pins_m_prod,
('NS in prod.' , 'Y10'): y10_pins_m_prod,
('NS out prod.' , 'Y09'): y09_pons_m_prod,
('NS out prod.' , 'Y10'): y10_pons_m_prod,
('NS in eco.' , 'Y09'): y09_pins_m_eco_prof,
('NS in eco.' , 'Y10'): y10_pins_m_eco_prof,
('NS out eco.' , 'Y09'): y09_pons_m_eco_prof,
('NS out eco.' , 'Y10'): y10_pons_m_eco_prof,
}
#
# both_year_comparison(my_d)
draw_month_change(my_d)
def run():
check_dir_create(summary_dir)
#
Y09_driver_total_monthly_fare, Y10_driver_total_monthly_fare = [], []
for y in xrange(9, 11):
for m in xrange(1, 13):
yymm = '%02d%02d' % (y, m)
if yymm in ['0912', '1010']:
continue
trip_df = pd.read_csv('%s/whole-trip-%s.csv' % (trips_dir, yymm))
trip_df = trip_df[(trip_df['did'] != -1)]
fares = [
x / CENT for x in list(trip_df.groupby(['did']).sum()['fare'])
]
if y == 9:
Y09_driver_total_monthly_fare += fares
else:
Y10_driver_total_monthly_fare += fares
save_pickle_file(
monthly_fare_summary,
[Y09_driver_total_monthly_fare, Y10_driver_total_monthly_fare])
def run():
check_dir_create(summary_dir)
#
cur_timestamp = datetime.datetime(2008, 12, 31, 23)
last_timestamp = datetime.datetime(2011, 1, 1, 0)
hp_summary, time_period_order = {}, []
while cur_timestamp < last_timestamp:
cur_timestamp += datetime.timedelta(hours=1)
yyyy, mm, dd, hh = cur_timestamp.year, cur_timestamp.month, cur_timestamp.day, cur_timestamp.hour
if yyyy == 2009 and mm == 12: continue
if yyyy == 2010 and mm == 10: continue
k = (str(yyyy - 2000), str(mm), str(dd), str(hh))
hp_summary[k] = [0 for _ in range(len([GEN_DUR, GEN_FARE, \
AP_DUR, AP_FARE, AP_QUEUE, \
NS_DUR, NS_FARE, NS_QUEUE]))]
time_period_order.append(k)
#
yy_l, mm_l, dd_l, hh_l = 'yy', 'mm', 'dd', 'hh'
# General
for fn in get_all_files(general_dur_fare_dir, general_dur_fare_prefix,
'.csv'):
print fn
with open('%s/%s' % (general_dur_fare_dir, fn), 'rb') as r_csvfile:
reader = csv.reader(r_csvfile)
headers = reader.next()
hid = {h: i for i, h in enumerate(headers)}
for row in reader:
yy, mm, dd, hh = row[hid[yy_l]], row[hid[mm_l]], row[
hid[dd_l]], row[hid[hh_l]]
k = (yy, mm, dd, hh)
if not hp_summary.has_key(k): continue
hp_summary[k][GEN_DUR] += eval(row[hid['gen-duration']])
hp_summary[k][GEN_FARE] += eval(row[hid['gen-fare']])
# Aiport
for fn in get_all_files(ap_dur_fare_q_time_dir, ap_dur_fare_q_time_prefix,
'.csv'):
print fn
with open('%s/%s' % (ap_dur_fare_q_time_dir, fn), 'rb') as r_csvfile:
reader = csv.reader(r_csvfile)
headers = reader.next()
hid = {h: i for i, h in enumerate(headers)}
for row in reader:
yy, mm, dd, hh = row[hid[yy_l]], row[hid[mm_l]], row[
hid[dd_l]], row[hid[hh_l]]
k = (yy, mm, dd, hh)
if not hp_summary.has_key(k): continue
hp_summary[k][AP_DUR] += eval(row[hid['ap-duration']])
hp_summary[k][AP_FARE] += eval(row[hid['ap-fare']])
hp_summary[k][AP_QUEUE] += eval(row[hid['ap-queue-time']])
# Night Safari
for fn in get_all_files(ns_dur_fare_q_time_dir, ns_dur_fare_q_time_prefix,
'.csv'):
print fn
with open('%s/%s' % (ns_dur_fare_q_time_dir, fn), 'rb') as r_csvfile:
reader = csv.reader(r_csvfile)
headers = reader.next()
hid = {h: i for i, h in enumerate(headers)}
for row in reader:
yy, mm, dd, hh = row[hid[yy_l]], row[hid[mm_l]], row[
hid[dd_l]], row[hid[hh_l]]
k = (yy, mm, dd, hh)
if not hp_summary.has_key(k): continue
hp_summary[k][NS_DUR] += eval(row[hid['ns-duration']])
hp_summary[k][NS_FARE] += eval(row[hid['ns-fare']])
hp_summary[k][NS_QUEUE] += eval(row[hid['ns-queue-time']])
# Summary
print 'summary'
zero_dur = []
with open(hourly_productivities, 'wt') as w_csvfile:
writer = csv.writer(w_csvfile)
header = [
'yy', 'mm', 'dd', 'hh', 'gen-duration', 'gen-fare', 'ap-duration',
'ap-fare', 'ap-queue-time', 'ns-duration', 'ns-fare',
'ns-queue-time', 'gen-productivity', 'ap-productivity',
'ap-out-productivity', 'ns-productivity', 'ns-out-productivity'
]
writer.writerow(header)
for k in time_period_order:
gen_dur, gen_fare, \
ap_dur, ap_fare, ap_queue, \
ns_dur, ns_fare, ns_queue = hp_summary[k]
yy, mm, dd, hh = k
#
try:
gen_prod = gen_fare / gen_dur
except ZeroDivisionError:
gen_prod = -1
zero_dur.append([GENERAL, k])
try:
ap_prod = ap_fare / (ap_dur + ap_queue)
except ZeroDivisionError:
ap_prod = -1
zero_dur.append([AIRPORT, k])
ap_out_prod = (gen_fare - ap_fare) / (gen_dur -
(ap_dur + ap_queue))
try:
ns_prod = ns_fare / (ns_dur + ns_queue)
except ZeroDivisionError:
ns_prod = -1
zero_dur.append([NIGHTSAFARI, k])
ns_out_prod = (gen_fare - ns_fare) / (gen_dur -
(ns_dur + ns_queue))
#
writer.writerow([
yy, mm, dd, hh, gen_dur, gen_fare, ap_dur, ap_fare, ap_queue,
ns_dur, ns_fare, ns_queue, gen_prod, ap_prod, ap_out_prod,
ns_prod, ns_out_prod
])
#
save_pickle_file(zero_duration_time_slots, zero_dur)
def run():
def difference(data0, data1):
diff = {}
for k, v in data0.iteritems():
diff[k] = data1[k] - v
return diff
check_dir_create(tables_dir)
write_text_file(ftd_group_analysis, 'Init', True)
#
# Load data_20160826
#
y09_gen_d_prod_for_ap, y10_gen_d_prod_for_ap = load_picle_file(
ftd_general_prod_db_for_ap)
y09_gen_d_prod_for_ns, y10_gen_d_prod_for_ns = load_picle_file(
ftd_general_prod_db_for_ns)
(y09_piap_d_prod, y09_piap_d_eco_prof), \
(y10_piap_d_prod, y10_piap_d_eco_prof), \
(y09_poap_d_prod, y09_poap_d_eco_prof), \
(y10_poap_d_prod, y10_poap_d_eco_prof) = load_picle_file(ftd_ap_prod_eco_prof_db)
(y09_pins_d_prod, y09_pins_d_eco_prof), \
(y10_pins_d_prod, y10_pins_d_eco_prof), \
(y09_pons_d_prod, y09_pons_d_eco_prof), \
(y10_pons_d_prod, y10_pons_d_eco_prof) = load_picle_file(ftd_ns_prod_eco_prof_db)
#
diffs_for_ap = {
'diff_g_prod_for_ap':
difference(y09_gen_d_prod_for_ap, y10_gen_d_prod_for_ap),
#
'diff_apin_prod':
difference(y09_piap_d_prod, y10_piap_d_prod),
'diff_apout_prod':
difference(y09_poap_d_prod, y10_poap_d_prod),
'diff_apin_eco':
difference(y09_piap_d_eco_prof, y10_piap_d_eco_prof),
'diff_apout_eco':
difference(y09_poap_d_eco_prof, y10_poap_d_eco_prof),
}
diffs_for_ns = {
'diff_g_prod_for_ns':
difference(y09_gen_d_prod_for_ns, y10_gen_d_prod_for_ns),
#
'diff_nsin_prod':
difference(y09_pins_d_prod, y10_pins_d_prod),
'diff_nsout_prod':
difference(y09_pons_d_prod, y10_pons_d_prod),
'diff_nsin_eco':
difference(y09_pins_d_eco_prof, y10_pins_d_eco_prof),
'diff_nsout_eco':
difference(y09_pons_d_eco_prof, y10_pons_d_eco_prof)
}
#
write_text_file(
ftd_group_analysis,
'Airport analysis ---------------------------------------------------')
an_analysis(diffs_for_ap)
write_text_file(
ftd_group_analysis,
'\nNight safari analysis ---------------------------------------------------'
)
ns_analysis(diffs_for_ns)
def run():
check_dir_create(summary_dir)
#
cur_timestamp = datetime.datetime(2008, 12, 31, 23)
last_timestamp = datetime.datetime(2011, 1, 1, 0)
hp_summary, time_period_order = {}, []
while cur_timestamp < last_timestamp:
cur_timestamp += datetime.timedelta(hours=1)
yyyy, mm, dd, hh = cur_timestamp.year, cur_timestamp.month, cur_timestamp.day, cur_timestamp.hour
if yyyy == 2009 and mm == 12: continue
if yyyy == 2010 and mm == 10: continue
k = (str(yyyy - 2000), str(mm), str(dd), str(hh))
hp_summary[k] = [0 for _ in range(len([GEN_DUR, GEN_FARE, \
AP_DUR, AP_FARE, AP_QUEUE, \
NS_DUR, NS_FARE, NS_QUEUE]))]
time_period_order.append(k)
#
yy_l, mm_l, dd_l, hh_l = 'yy', 'mm', 'dd', 'hh'
# General
for fn in get_all_files(general_dur_fare_dir, general_dur_fare_prefix, '.csv'):
print fn
with open('%s/%s' % (general_dur_fare_dir, fn), 'rb') as r_csvfile:
reader = csv.reader(r_csvfile)
headers = reader.next()
hid = {h : i for i, h in enumerate(headers)}
for row in reader:
yy, mm, dd, hh = row[hid[yy_l]], row[hid[mm_l]], row[hid[dd_l]], row[hid[hh_l]]
k = (yy, mm, dd, hh)
if not hp_summary.has_key(k): continue
hp_summary[k][GEN_DUR] += eval(row[hid['gen-duration']])
hp_summary[k][GEN_FARE] += eval(row[hid['gen-fare']])
# Aiport
for fn in get_all_files(ap_dur_fare_q_time_dir, ap_dur_fare_q_time_prefix, '.csv'):
print fn
with open('%s/%s' % (ap_dur_fare_q_time_dir, fn), 'rb') as r_csvfile:
reader = csv.reader(r_csvfile)
headers = reader.next()
hid = {h : i for i, h in enumerate(headers)}
for row in reader:
yy, mm, dd, hh = row[hid[yy_l]], row[hid[mm_l]], row[hid[dd_l]], row[hid[hh_l]]
k = (yy, mm, dd, hh)
if not hp_summary.has_key(k): continue
hp_summary[k][AP_DUR] += eval(row[hid['ap-duration']])
hp_summary[k][AP_FARE] += eval(row[hid['ap-fare']])
hp_summary[k][AP_QUEUE] += eval(row[hid['ap-queue-time']])
# Night Safari
for fn in get_all_files(ns_dur_fare_q_time_dir, ns_dur_fare_q_time_prefix, '.csv'):
print fn
with open('%s/%s' % (ns_dur_fare_q_time_dir, fn), 'rb') as r_csvfile:
reader = csv.reader(r_csvfile)
headers = reader.next()
hid = {h : i for i, h in enumerate(headers)}
for row in reader:
yy, mm, dd, hh = row[hid[yy_l]], row[hid[mm_l]], row[hid[dd_l]], row[hid[hh_l]]
k = (yy, mm, dd, hh)
if not hp_summary.has_key(k): continue
hp_summary[k][NS_DUR] += eval(row[hid['ns-duration']])
hp_summary[k][NS_FARE] += eval(row[hid['ns-fare']])
hp_summary[k][NS_QUEUE] += eval(row[hid['ns-queue-time']])
# Summary
print 'summary'
zero_dur = []
with open(hourly_productivities, 'wt') as w_csvfile:
writer = csv.writer(w_csvfile)
header = ['yy', 'mm', 'dd', 'hh',
'gen-duration', 'gen-fare',
'ap-duration', 'ap-fare', 'ap-queue-time',
'ns-duration', 'ns-fare', 'ns-queue-time',
'gen-productivity',
'ap-productivity', 'ap-out-productivity',
'ns-productivity', 'ns-out-productivity']
writer.writerow(header)
for k in time_period_order:
gen_dur, gen_fare, \
ap_dur, ap_fare, ap_queue, \
ns_dur, ns_fare, ns_queue = hp_summary[k]
yy, mm, dd, hh = k
#
try:
gen_prod = gen_fare / gen_dur
except ZeroDivisionError:
gen_prod = -1
zero_dur.append([GENERAL, k])
try:
ap_prod = ap_fare / (ap_dur + ap_queue)
except ZeroDivisionError:
ap_prod = -1
zero_dur.append([AIRPORT, k])
ap_out_prod = (gen_fare - ap_fare) / (gen_dur - (ap_dur + ap_queue))
try:
ns_prod = ns_fare / (ns_dur + ns_queue)
except ZeroDivisionError:
ns_prod = -1
zero_dur.append([NIGHTSAFARI, k])
ns_out_prod = (gen_fare - ns_fare) / (gen_dur - (ns_dur + ns_queue))
#
writer.writerow([yy, mm, dd, hh,
gen_dur, gen_fare,
ap_dur, ap_fare, ap_queue,
ns_dur, ns_fare, ns_queue,
gen_prod,
ap_prod, ap_out_prod,
ns_prod, ns_out_prod])
#
save_pickle_file(zero_duration_time_slots, zero_dur)