Note

This page was generated from user_guide/data_analysis/dlr-ut-dataset.pct.py.
Interactive online version: Binder badge

Using the DLRUTDataset#

This example shall give an overview of the methods and attributes that are available in the DLRUTDataset class.

Load trajectory data#

At first, we need to load the trajectory data of the dataset.

[1]:

from tasi.dlr.dataset import DLRUTDatasetManager, DLRUTVersion
from tasi.dlr import DLRTrajectoryDataset

dataset = DLRUTDatasetManager(DLRUTVersion.latest)
dataset.load()

ds = DLRTrajectoryDataset.from_csv(dataset.trajectory()[50])

[2025-04-25 12:46:34 | dataset.py:load:112] > INFO:  Checking if dataset already downloaded /tmp/DLR-Urban-Traffic-dataset_v1-2-0
[2025-04-25 12:46:34 | dataset.py:load:138] > INFO:  Dataset already available at /tmp/DLR-Urban-Traffic-dataset_v1-2-0

Attributes of the dataset#

There are several attributes available to get information about a dataset. For instance, we can get the interval of a dataset via the property

[2]:

ds.interval

[2]:

Interval(2023-09-24 12:30:00.016482+00:00, 2023-09-24 12:44:59.966482+00:00, closed='right')

or all unique timestamps of it via

[3]:

ds.timestamps

[3]:

DatetimeIndex(['2023-09-24 12:30:00.016482+00:00',
               '2023-09-24 12:30:00.066482+00:00',
               '2023-09-24 12:30:00.116482+00:00',
               '2023-09-24 12:30:00.166482+00:00',
               '2023-09-24 12:30:00.216482+00:00',
               '2023-09-24 12:30:00.266482+00:00',
               '2023-09-24 12:30:00.316482+00:00',
               '2023-09-24 12:30:00.366482+00:00',
               '2023-09-24 12:30:00.416482+00:00',
               '2023-09-24 12:30:00.466482+00:00',
               ...
               '2023-09-24 12:44:59.516482+00:00',
               '2023-09-24 12:44:59.566482+00:00',
               '2023-09-24 12:44:59.616482+00:00',
               '2023-09-24 12:44:59.666482+00:00',
               '2023-09-24 12:44:59.716482+00:00',
               '2023-09-24 12:44:59.766482+00:00',
               '2023-09-24 12:44:59.816482+00:00',
               '2023-09-24 12:44:59.866482+00:00',
               '2023-09-24 12:44:59.916482+00:00',
               '2023-09-24 12:44:59.966482+00:00'],
              dtype='datetime64[ns, UTC]', name='timestamp', length=18000, freq=None)

or the ids of all traffic participants in the dataset.

[4]:

ds.ids

[4]:

Index([1695558514712340, 1695558558975558, 1695558561769745, 1695558563216597,
       1695558565170472, 1695558571572035, 1695558572272810, 1695558573520980,
       1695558576618026, 1695558579216721,
       ...
       1695559487523035, 1695559488624461, 1695559490123751, 1695559490123748,
       1695559491223905, 1695559492674737, 1695559496120575, 1695559498274800,
       1695559499574101, 1695559499171792],
      dtype='int64', name='id', length=750)

Filtering#

If you want to look into a short sequence of the overall dataset, you can select specific rows of the overall dataset. The DLRTrajectoryDataset provides various ways for this purpose.

Time and object#

There are two variants to filter a dataset based on the information on the dataset’s index. For instance, if you want to filterthe dataset by an interval, you can utilize the during method

[5]:

ds.during(ds.timestamps[0], ds.timestamps[10])

[5]:
center velocity acceleration yaw dimension classifications interpolated
easting northing easting northing magnitude easting northing magnitude length width height pedestrian bicycle motorbike car van truck
timestamp id
2023-09-24 12:30:00.016482+00:00 1695558514712340 604774.496 5792796.320 7.140 -3.970 8.169 1.068 1.320 1.698 -29.398 2.536 1.177 1.445 0.000 0.008 0.388 0.598 0.0 0.006 False
1695558558975558 604740.884 5792779.240 0.021 0.000 0.021 0.014 -0.008 0.016 7.852 2.669 1.259 1.261 0.000 0.000 0.163 0.838 0.0 0.000 False
1695558561769745 604787.539 5792764.373 -0.008 0.041 0.042 0.008 -0.035 0.036 107.830 3.445 1.409 1.311 0.011 0.000 0.037 0.952 0.0 0.000 False
1695558563216597 604741.853 5792776.267 -0.030 0.032 0.044 -0.041 -0.014 0.044 9.183 3.323 1.293 1.468 0.000 0.000 0.026 0.954 0.0 0.000 False
1695558565170472 604791.430 5792770.438 -0.000 0.013 0.013 -0.002 -0.004 0.004 116.146 3.407 1.775 1.556 0.000 0.000 0.000 1.000 0.0 0.000 False
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
2023-09-24 12:30:00.466482+00:00 1695558592618702 604725.976 5792780.341 3.969 0.516 4.002 -1.060 -0.095 1.065 7.403 3.387 1.496 1.496 0.000 0.000 0.095 0.904 0.0 0.001 False
1695558596566965 604744.173 5792789.955 0.441 -1.464 1.529 0.117 -0.155 0.194 -73.101 1.323 0.787 1.624 0.837 0.163 0.000 0.000 0.0 0.000 False
1695558596716951 604742.064 5792846.878 0.674 -2.246 2.345 0.054 -0.237 0.243 -73.240 2.848 1.394 1.368 0.000 0.000 0.062 0.933 0.0 0.000 False
1695558598618723 604745.924 5792793.615 1.345 -2.617 2.942 -0.116 0.072 0.137 -62.834 0.939 1.029 1.445 0.000 1.000 0.000 0.000 0.0 0.000 False
1695558598820336 604747.604 5792840.329 3.606 -10.172 10.793 0.091 -0.122 0.152 -70.505 2.950 1.698 1.640 0.000 0.000 0.068 0.922 0.0 0.010 False

172 rows × 19 columns

that returns the rows within the given interval.

Another variant to select specific rows of the datasets is by the id of a traffic participant. This might be useful if you want to take a closer look into the behavior of specific traffic participants. For instance, to filter by the second traffic participant in the dataset, we can combine the ids attribute with the trajectory method.

[6]:

ds.trajectory(ds.ids[1])

[6]:
center velocity acceleration yaw dimension classifications interpolated
easting northing easting northing magnitude easting northing magnitude length width height pedestrian bicycle motorbike car van truck
timestamp id
2023-09-24 12:30:00.016482+00:00 1695558558975558 604740.884 5792779.240 0.021 0.000 0.021 0.014 -0.008 0.016 7.852 2.669 1.259 1.261 0.0 0.0 0.163 0.838 0.0 0.0 False
2023-09-24 12:30:00.066482+00:00 1695558558975558 604740.884 5792779.240 0.023 -0.001 0.023 0.013 -0.008 0.015 7.852 2.669 1.259 1.261 0.0 0.0 0.163 0.838 0.0 0.0 False
2023-09-24 12:30:00.116482+00:00 1695558558975558 604740.884 5792779.240 0.025 -0.001 0.025 0.012 -0.008 0.014 7.852 2.669 1.259 1.261 0.0 0.0 0.163 0.838 0.0 0.0 False
2023-09-24 12:30:00.166482+00:00 1695558558975558 604740.884 5792779.240 0.028 -0.002 0.028 0.011 -0.007 0.014 7.852 2.669 1.259 1.261 0.0 0.0 0.163 0.838 0.0 0.0 False
2023-09-24 12:30:00.216482+00:00 1695558558975558 604740.885 5792779.240 0.030 -0.002 0.030 0.010 -0.007 0.013 7.852 2.669 1.259 1.261 0.0 0.0 0.163 0.838 0.0 0.0 False
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
2023-09-24 12:30:34.216482+00:00 1695558558975558 604783.725 5792742.854 3.161 -9.792 10.290 -0.242 -0.183 0.304 -72.008 2.669 1.259 1.261 0.0 0.0 0.163 0.838 0.0 0.0 False
2023-09-24 12:30:34.266482+00:00 1695558558975558 604783.882 5792742.364 3.153 -9.798 10.293 -0.242 -0.183 0.304 -72.049 2.669 1.259 1.261 0.0 0.0 0.163 0.838 0.0 0.0 False
2023-09-24 12:30:34.316482+00:00 1695558558975558 604784.038 5792741.873 3.146 -9.803 10.296 -0.242 -0.183 0.304 -72.084 2.669 1.259 1.261 0.0 0.0 0.163 0.838 0.0 0.0 False
2023-09-24 12:30:34.366482+00:00 1695558558975558 604784.194 5792741.382 3.140 -9.808 10.298 -0.242 -0.183 0.304 -72.114 2.669 1.259 1.261 0.0 0.0 0.163 0.838 0.0 0.0 False
2023-09-24 12:30:34.416482+00:00 1695558558975558 604784.349 5792740.890 3.134 -9.812 10.301 -0.242 -0.183 0.304 -72.141 2.669 1.259 1.261 0.0 0.0 0.163 0.838 0.0 0.0 False

689 rows × 19 columns

Traffic participant properties#

There are also methods available that might help to find the relevant information in the dataset. The most straight forward option is to use pandas’ capability to access specific attributesof the datasets. The available attributes on the dataset, are available via theattribute` property.

[7]:

ds.attributes

[7]:

Index(['center', 'velocity', 'acceleration', 'yaw', 'dimension',
       'classifications', 'interpolated'],
      dtype='object')

We can, for instance, access the traffic participants center position.

[8]:

ds.center

[8]:
easting northing
timestamp id
2023-09-24 12:30:00.016482+00:00 1695558514712340 604774.496 5792796.320
1695558558975558 604740.884 5792779.240
1695558561769745 604787.539 5792764.373
1695558563216597 604741.853 5792776.267
1695558565170472 604791.430 5792770.438
... ... ... ...
2023-09-24 12:44:59.966482+00:00 1695559492674737 604789.200 5792769.423
1695559496120575 604702.210 5792771.596
1695559498274800 604855.834 5792822.882
1695559499171792 604740.345 5792785.036
1695559499574101 604871.056 5792836.959

388075 rows × 2 columns

or the classification propabilities.

[9]:

ds.classifications

[9]:
pedestrian bicycle motorbike car van truck
timestamp id
2023-09-24 12:30:00.016482+00:00 1695558514712340 0.000 0.008 0.388 0.598 0.0 0.006
1695558558975558 0.000 0.000 0.163 0.838 0.0 0.000
1695558561769745 0.011 0.000 0.037 0.952 0.0 0.000
1695558563216597 0.000 0.000 0.026 0.954 0.0 0.000
1695558565170472 0.000 0.000 0.000 1.000 0.0 0.000
... ... ... ... ... ... ... ...
2023-09-24 12:44:59.966482+00:00 1695559492674737 0.000 0.000 0.021 0.979 0.0 0.000
1695559496120575 0.001 0.014 0.414 0.572 0.0 0.000
1695559498274800 0.000 0.000 0.079 0.921 0.0 0.000
1695559499171792 0.000 0.000 0.340 0.660 0.0 0.000
1695559499574101 0.000 0.022 0.089 0.890 0.0 0.000

388075 rows × 6 columns

We extended these basic capabilities with additional methods, that, for instance, allow to get the most likely class by each traffic participant’s pose

[10]:

ds.most_likely_class(by='pose')

[10]:

timestamp                         id
2023-09-24 12:30:00.016482+00:00  1695558514712340    car
                                  1695558558975558    car
                                  1695558561769745    car
                                  1695558563216597    car
                                  1695558565170472    car
                                                     ...
2023-09-24 12:44:59.966482+00:00  1695559492674737    car
                                  1695559496120575    car
                                  1695559498274800    car
                                  1695559499171792    car
                                  1695559499574101    car
Length: 388075, dtype: object

or by the overall trajectory (the default), i.e. all poses of a traffic participants.

[11]:

ds.most_likely_class(by='trajectory')

[11]:

id
1695558514712340    car
1695558558975558    car
1695558561769745    car
1695558563216597    car
1695558565170472    car
                   ...
1695559492674737    car
1695559496120575    car
1695559498274800    car
1695559499171792    car
1695559499574101    car
Length: 750, dtype: object

This might help to filter the dataset to select only traffic participants that are classified as a car. To archieve this, we first get the most likely class per trajectory, select the rows having the value ‘car’ and pass their index (the traffic particpant’s id) into the trajectory method of the Dataset.

[12]:

classification = ds.most_likely_class(by='trajectory')

ds.trajectory(classification[classification == 'car'].index)

[12]:
center velocity acceleration yaw dimension classifications interpolated
easting northing easting northing magnitude easting northing magnitude length width height pedestrian bicycle motorbike car van truck
timestamp id
2023-09-24 12:30:00.016482+00:00 1695558514712340 604774.496 5792796.320 7.140 -3.970 8.169 1.068 1.320 1.698 -29.398 2.536 1.177 1.445 0.000 0.008 0.388 0.598 0.0 0.006 False
1695558558975558 604740.884 5792779.240 0.021 0.000 0.021 0.014 -0.008 0.016 7.852 2.669 1.259 1.261 0.000 0.000 0.163 0.838 0.0 0.000 False
1695558561769745 604787.539 5792764.373 -0.008 0.041 0.042 0.008 -0.035 0.036 107.830 3.445 1.409 1.311 0.011 0.000 0.037 0.952 0.0 0.000 False
1695558563216597 604741.853 5792776.267 -0.030 0.032 0.044 -0.041 -0.014 0.044 9.183 3.323 1.293 1.468 0.000 0.000 0.026 0.954 0.0 0.000 False
1695558565170472 604791.430 5792770.438 -0.000 0.013 0.013 -0.002 -0.004 0.004 116.146 3.407 1.775 1.556 0.000 0.000 0.000 1.000 0.0 0.000 False
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
2023-09-24 12:44:59.966482+00:00 1695559492674737 604789.200 5792769.423 -0.813 2.471 2.602 0.612 -2.115 2.202 108.191 3.642 1.634 1.318 0.000 0.000 0.021 0.979 0.0 0.000 False
1695559496120575 604702.210 5792771.596 8.734 0.456 8.745 -0.850 0.111 0.857 2.955 2.280 0.931 1.304 0.001 0.014 0.414 0.572 0.0 0.000 False
1695559498274800 604855.834 5792822.882 -13.271 -5.832 14.496 -0.009 0.252 0.253 -156.208 3.026 1.385 1.385 0.000 0.000 0.079 0.921 0.0 0.000 False
1695559499171792 604740.345 5792785.036 5.195 0.908 5.274 1.309 0.118 1.314 9.847 2.907 1.451 1.269 0.000 0.000 0.340 0.660 0.0 0.000 False
1695559499574101 604871.056 5792836.959 -15.391 -5.728 16.423 -0.422 -0.030 0.423 -159.568 3.028 1.593 1.389 0.000 0.022 0.089 0.890 0.0 0.000 False

344669 rows × 19 columns

On this page
Show Source