def test_aggregating_further_by_station_name_with_missing_names():
df = DataFrame({
'station_A': {
0: '4',
5: '01'
},
'station_B': {
0: '1',
5: '04'
},
'station_A_name': {
0: '',
5: 'Stop_1'
},
'station_B_name': {
0: 'Stop_1',
5: ''
},
'mode': {
0: 'bus',
5: 'bus'
},
'routes_in_common': {
0: {'11', '1'},
5: {'012', '2'}
},
'number_of_trips': {
0: 3,
5: 3
}
})
aggregated = use_schedule.aggregate_by_stop_names(df)
assert aggregated.empty
def test_aggregating_trips_per_day_per_route_by_end_stop_pairs_with_different_modes():
df = DataFrame(
{'station_A': {0: '4', 1: '01', 2: 'a', 3: 'b'},
'station_B': {0: '1', 1: '04', 2: 'b', 3: 'a'},
'station_A_name': {0: 'Stop_4', 1: 'Stop_1', 2: 'A', 3: 'B'},
'station_B_name': {0: 'Stop_1', 1: 'Stop_4', 2: 'B', 3: 'A'},
'mode': {0: 'bus', 1: 'bus', 2: 'rail', 3: 'horse'},
'routes_in_common': {0: {'11', '1'}, 1: {'012', '2'}, 2: {'lol'}, 3: {''}},
'number_of_trips': {0: 3, 1: 3, 2: 4, 3: 10}})
aggregated = use_schedule.aggregate_by_stop_names(df)
assert_frame_equal(aggregated, DataFrame(
{'station_A_name': {0: 'A', 1: 'A', 2: 'Stop_1'},
'station_B_name': {0: 'B', 1: 'B', 2: 'Stop_4'},
'mode': {0: 'horse', 1: 'rail', 2: 'bus'},
'number_of_trips': {0: 10, 1: 4, 2: 6}}))
def generate_standard_outputs_for_schedule(schedule, output_dir, gtfs_day='19700101', include_shp_files=False):
logging.info('Generating geojson standard outputs for schedule')
schedule_links = schedule.to_geodataframe()['links'].to_crs("epsg:4326")
df = schedule.route_trips_with_stops_to_dataframe(gtfs_day=gtfs_day)
df_all_modes_vph = None
vph_dir = os.path.join(output_dir, 'vehicles_per_hour')
subgraph_dir = os.path.join(output_dir, 'subgraphs')
graph_mode_map = schedule.mode_graph_map()
for mode in schedule.modes():
logging.info(f'Generating vehicles per hour for {mode}')
df_vph = use_schedule.generate_edge_vph_geodataframe(df[df['mode'] == mode], schedule_links)
save_geodataframe(
df_vph,
filename=f'vehicles_per_hour_{mode}',
output_dir=vph_dir,
include_shp_files=include_shp_files
)
if df_all_modes_vph is None:
df_vph['mode'] = mode
df_all_modes_vph = df_vph
else:
df_vph['mode'] = mode
df_all_modes_vph = df_all_modes_vph.append(df_vph)
logging.info(f'Generating schedule graph for {mode}')
schedule_subgraph = generate_geodataframes(
schedule.subgraph(graph_mode_map[mode]))
save_geodataframe(
schedule_subgraph['links'].to_crs("epsg:4326"),
filename=f'schedule_subgraph_links_{mode}',
output_dir=subgraph_dir,
include_shp_files=include_shp_files
)
save_geodataframe(
schedule_subgraph['nodes'].to_crs("epsg:4326"),
filename=f'schedule_subgraph_nodes_{mode}',
output_dir=subgraph_dir,
include_shp_files=include_shp_files
)
logging.info('Saving vehicles per hour for all PT modes')
save_geodataframe(
df_all_modes_vph,
filename='vehicles_per_hour_all_modes',
output_dir=vph_dir,
include_shp_files=include_shp_files
)
logging.info('Saving vehicles per hour for all PT modes for selected hour slices')
for h in [7, 8, 9, 13, 16, 17, 18]:
save_geodataframe(
df_all_modes_vph[df_all_modes_vph['hour'].dt.hour == h],
filename=f'vph_all_modes_within_{h-1}:30-{h}:30',
output_dir=vph_dir,
include_shp_files=include_shp_files
)
logging.info('Generating csv for vehicles per hour for each service')
use_schedule.vehicles_per_hour(
df,
aggregate_by=['service', 'service_name', 'mode'],
output_path=os.path.join(vph_dir, 'vph_per_service.csv'))
logging.info('Generating csv for vehicles per hour per stop')
use_schedule.vehicles_per_hour(
df,
aggregate_by=['from_stop', 'from_stop_name', 'mode'],
output_path=os.path.join(vph_dir, 'vph_per_stop_departing_from.csv'))
use_schedule.vehicles_per_hour(
df,
aggregate_by=['to_stop', 'to_stop_name', 'mode'],
output_path=os.path.join(vph_dir, 'vph_per_stop_arriving_at.csv'))
logging.info('Generating csvs for trips per day')
use_schedule.trips_per_day_per_service(df, output_dir=output_dir)
df_trips_per_route = use_schedule.trips_per_day_per_route(df, output_dir=output_dir)
# stop-to-stop trips per day aggregation
aggregated_per_stops = use_schedule.aggregate_trips_per_day_per_route_by_end_stop_pairs(
schedule, df_trips_per_route)
aggregated_per_stops.to_csv(os.path.join(output_dir, 'trips_per_day_per_route_aggregated_per_stop_id_pair.csv'))
use_schedule.aggregate_by_stop_names(aggregated_per_stops).to_csv(
os.path.join(output_dir, 'trips_per_day_per_route_aggregated_per_stop_name_pair.csv'))
