Point Sampling¶

MATSim requires point locations for all activities and trips. More generally it can be useful to assign geometries to activity location and/or trip origin and destinations. This notebook shows how to sample geometric locations from ozone/dzone regions using a simple random point sampler.

More advanced samplers are available in pam.samplers.spatial.

In [1]:

import os

import geopandas as gp
import pandas as pd
from matplotlib import pyplot as plt
from pam import read
from pam.samplers.spatial import RandomPointSampler

import os import geopandas as gp import pandas as pd from matplotlib import pyplot as plt from pam import read from pam.samplers.spatial import RandomPointSampler

In [2]:

trips = pd.read_csv(
    os.path.join("data", "example_data", "example_travel_diaries.csv"), index_col="uid"
)
attributes = pd.read_csv(
    os.path.join("data", "example_data", "example_attributes.csv"), index_col="pid"
)
trips.head(10)

trips = pd.read_csv( os.path.join("data", "example_data", "example_travel_diaries.csv"), index_col="uid" ) attributes = pd.read_csv( os.path.join("data", "example_data", "example_attributes.csv"), index_col="pid" ) trips.head(10)

Out[2]:

	pid	hid	seq	hzone	ozone	dzone	purp	mode	tst	tet	freq
uid
0	census_0	census_0	0	Harrow	Harrow	Camden	work	pt	444	473	1000
1	census_0	census_0	1	Harrow	Camden	Harrow	work	pt	890	919	1000
2	census_1	census_1	0	Greenwich	Greenwich	Tower Hamlets	work	pt	507	528	1000
3	census_1	census_1	1	Greenwich	Tower Hamlets	Greenwich	work	pt	1065	1086	1000
4	census_2	census_2	0	Croydon	Croydon	Croydon	work	pt	422	425	1000
5	census_2	census_2	1	Croydon	Croydon	Croydon	work	pt	917	920	1000
6	census_3	census_3	0	Haringey	Haringey	Redbridge	work	pt	428	447	1000
7	census_3	census_3	1	Haringey	Redbridge	Haringey	work	pt	1007	1026	1000
8	census_4	census_4	0	Hounslow	Hounslow	Westminster,City of London	work	car	483	516	1000
9	census_4	census_4	1	Hounslow	Westminster,City of London	Hounslow	work	car	1017	1050	1000

In [3]:

population = read.load_travel_diary(trips, attributes)
population.stats

population = read.load_travel_diary(trips, attributes) population.stats

Using tour based purpose parser (recommended)

Adding pid->hh mapping to persons_attributes from trips.

Adding home locations to persons attributes using trips attributes.

Using freq of 'None' for all persons.

Out[3]:

{'num_households': 51,
 'num_people': 51,
 'num_activities': 152,
 'num_legs': 101}

Sample Geometries¶

Transform trip ozone/dzone to geographically sampled points

In [4]:

zones = gp.read_file(os.path.join("data", "example_data", "geometry.geojson")).set_index("NAME")
zones.head()

zones = gp.read_file(os.path.join("data", "example_data", "geometry.geojson")).set_index("NAME") zones.head()

Out[4]:

	HECTARES	NONLD_AREA	ONS_INNER	SUB_2009	SUB_2006	geometry
NAME
Kingston upon Thames	3726.117	0.000	F	None	None	POLYGON ((516401.600 160201.800, 516407.300 16...
Croydon	8649.441	0.000	F	None	None	POLYGON ((535009.200 159504.700, 535005.500 15...
Bromley	15013.487	0.000	F	None	None	POLYGON ((540373.600 157530.400, 540361.200 15...
Hounslow	5658.541	60.755	F	None	None	POLYGON ((521975.800 178100.000, 521967.700 17...
Ealing	5554.428	0.000	F	None	None	POLYGON ((510253.500 182881.600, 510249.900 18...

In [5]:

random_sampler = RandomPointSampler(geoms=zones)

random_sampler = RandomPointSampler(geoms=zones)

In [6]:

# test
fig, ax = plt.subplots(1)
zones.plot(ax=ax, color="grey")
zones.loc[["Camden"]].plot(ax=ax)
point = random_sampler.sample("Camden", activity=None)
ax.scatter(point.x, point.y, c="red")

# test fig, ax = plt.subplots(1) zones.plot(ax=ax, color="grey") zones.loc[["Camden"]].plot(ax=ax) point = random_sampler.sample("Camden", activity=None) ax.scatter(point.x, point.y, c="red")

Out[6]:

<matplotlib.collections.PathCollection at 0x15f0627a0>

In [7]:

population.sample_locs(random_sampler)  # apply to whole population

population.sample_locs(random_sampler) # apply to whole population

Let's check out an example Activity Plan and Attributes:

In [8]:

person = population["census_0"]["census_0"]
person.print()
person.plot()

person = population["census_0"]["census_0"] person.print() person.plot()

Person: census_0
{'gender': 'female', 'job': 'work', 'occ': 'white', 'inc': 'low', 'hzone': 'Harrow'}
0:	Activity(act:home, location:POINT (516104.2142596701 187104.23657894367), time:00:00:00 --> 07:24:00, duration:7:24:00)
1:	Leg(mode:pt, area:POINT (516104.2142596701 187104.23657894367) --> POINT (528574.2978802138 186510.35985765653), time:07:24:00 --> 07:53:00, duration:0:29:00)
2:	Activity(act:work, location:POINT (528574.2978802138 186510.35985765653), time:07:53:00 --> 14:50:00, duration:6:57:00)
3:	Leg(mode:pt, area:POINT (528574.2978802138 186510.35985765653) --> POINT (516104.2142596701 187104.23657894367), time:14:50:00 --> 15:19:00, duration:0:29:00)
4:	Activity(act:home, location:POINT (516104.2142596701 187104.23657894367), time:15:19:00 --> 00:00:00, duration:8:41:00)