Chicago crime

Chicago Crime Dataset¶

This dataset reflects reported incidents of crime (with the exception of murders where data exists for each victim) that occurred in the City of Chicago. Data is extracted from the Chicago Police Department's CLEAR (Citizen Law Enforcement Analysis and Reporting) system. Each entry includes precise time and location details, supporting fine-grained spatial and temporal analysis. For benchmarking, a subset of reports from 2022 is used, while raw multi-year data is also provided without predefined train–test splits.

Run this notebook in Google Colab:

Remember to install the srai library before running the notebook:

%pip install srai[all]

In [1]:

Copied!





# plotting imports
import contextily as cx
import numpy as np
import seaborn as sns
from matplotlib import pyplot as plt
from matplotlib.patches import Patch

# dataset import
from srai.datasets import ChicagoCrimeDataset
# plotting imports
import contextily as cx
import numpy as np
import seaborn as sns
from matplotlib import pyplot as plt
from matplotlib.patches import Patch

# dataset import
from srai.datasets import ChicagoCrimeDataset

In [2]:

Copied!

chicago_crime = ChicagoCrimeDataset()
chicago_crime = ChicagoCrimeDataset()

Load default data

In [3]:

Copied!

ds = chicago_crime.load()
ds.keys()
ds = chicago_crime.load()
ds.keys()

Out[3]:

dict_keys(['train', 'test'])

In [4]:

Copied!

type(chicago_crime.train_gdf), type(chicago_crime.test_gdf)
type(chicago_crime.train_gdf), type(chicago_crime.test_gdf)

Out[4]:

(geopandas.geodataframe.GeoDataFrame, geopandas.geodataframe.GeoDataFrame)

In [5]:

Copied!

print("Aggregation H3 resolution:", chicago_crime.resolution)
print("Aggregation H3 resolution:", chicago_crime.resolution)

Aggregation H3 resolution: 9

In [6]:

Copied!

print("Prediction target:", chicago_crime.target)
print("Prediction target:", chicago_crime.target)

Prediction target: count

In [7]:

Copied!

gdf_train, gdf_test = ds["train"], ds["test"]
gdf_train, gdf_test = ds["train"], ds["test"]

In [8]:

Copied!

gdf_train.head()
gdf_train.head()

Out[8]:

	ID	Case Number	Date	Block	IUCR	Primary Type	Description	Location Description	Arrest	Domestic	Beat	District	Ward	Community Area	FBI Code	Year	Updated On	geometry
0	12589893	JF109865	01/11/2022 03:00:00 PM	087XX S KINGSTON AVE	1565	SEX OFFENSE	INDECENT SOLICITATION OF A CHILD	RESIDENCE	False	True	423	4	7.0	46.0	17	2022	09/14/2023 03:41:59 PM	POINT (-87.56241 41.73641)
1	12592454	JF113025	01/14/2022 03:55:00 PM	067XX S MORGAN ST	2826	OTHER OFFENSE	HARASSMENT BY ELECTRONIC MEANS	RESIDENCE	False	True	724	7	16.0	68.0	26	2022	09/14/2023 03:41:59 PM	POINT (-87.64944 41.77178)
2	12785595	JF346553	08/05/2022 09:00:00 PM	072XX S UNIVERSITY AVE	1544	SEX OFFENSE	SEXUAL EXPLOITATION OF A CHILD	APARTMENT	True	False	324	3	5.0	69.0	17	2022	09/14/2023 03:41:59 PM	POINT (-87.597 41.76334)
3	12808281	JF373517	08/14/2022 02:00:00 PM	055XX W ARDMORE AVE	1562	SEX OFFENSE	AGGRAVATED CRIMINAL SEXUAL ABUSE	RESIDENCE	False	False	1621	16	39.0	11.0	17	2022	09/14/2023 03:41:59 PM	POINT (-87.7664 41.98588)
4	12888104	JF469015	11/10/2022 03:47:00 AM	072XX S MAY ST	1477	WEAPONS VIOLATION	RECKLESS FIREARM DISCHARGE	STREET	False	False	733	7	17.0	68.0	15	2022	09/14/2023 03:41:59 PM	POINT (-87.65284 41.76261)

Getting target values for h3

In [9]:

Copied!

train_h3, _, test_h3 = chicago_crime.get_h3_with_labels()
train_h3, _, test_h3 = chicago_crime.get_h3_with_labels()

In [10]:

Copied!

train_h3.head()
train_h3.head()

Out[10]:

	geometry	count
region_id
892664cc143ffff	POLYGON ((-87.61025 41.76848, -87.61206 41.767...	0.104348
892664cce0bffff	POLYGON ((-87.61168 41.74644, -87.61349 41.745...	0.094410
892664cb063ffff	POLYGON ((-87.80618 41.94159, -87.80799 41.940...	0.034783
892664ca613ffff	POLYGON ((-87.73913 41.95981, -87.74095 41.958...	0.007453
892664cdeb7ffff	POLYGON ((-87.67165 41.75393, -87.67346 41.752...	0.099379

In [11]:

Copied!

test_h3.head()
test_h3.head()

Out[11]:

	geometry	count
region_id
892664ccd27ffff	POLYGON ((-87.54947 41.7256, -87.55128 41.7244...	0.008696
892664cb303ffff	POLYGON ((-87.80807 41.92763, -87.80989 41.926...	0.003727
892664c8883ffff	POLYGON ((-87.72613 41.84962, -87.72795 41.848...	0.047205
892664cc437ffff	POLYGON ((-87.60214 41.79698, -87.60395 41.795...	0.027329
892664562d3ffff	POLYGON ((-87.66622 41.69198, -87.66803 41.690...	0.027329

In [12]:

Copied!





fig, axes = plt.subplots(
    2, 1, sharex=False, sharey=False, figsize=(12, 19), height_ratios=[4, 1]
)

train_h3.plot(
    color="orange",
    markersize=0.1,
    ax=axes[0],
    label="train",
    alpha=np.minimum(np.power(train_h3[chicago_crime.target] + 0.4, 2), 1),
)
test_h3.plot(
    color="royalblue",
    markersize=0.1,
    ax=axes[0],
    label="test",
    alpha=np.minimum(np.power(test_h3[chicago_crime.target] + 0.4, 2), 1),
)

cx.add_basemap(axes[0], source=cx.providers.CartoDB.PositronNoLabels, crs=4326, zoom=12)
axes[0].set_title("Chicago crime data aggregated to H3 cells")
axes[0].legend(
    handles=[Patch(facecolor="orange"), Patch(facecolor="royalblue")],
    labels=["Train", "Test"],
)
axes[0].set_axis_off()

sns.kdeplot(
    x=train_h3[chicago_crime.target],
    label="train",
    color="orange",
    ax=axes[1],
    fill=False,
    cut=0,
)
sns.kdeplot(
    x=test_h3[chicago_crime.target],
    label="test",
    color="royalblue",
    ax=axes[1],
    fill=False,
    cut=0,
)
axes[1].set_title("Chicago crime data - target distribution")
axes[1].legend()

fig.tight_layout()

plt.show()
fig, axes = plt.subplots(
    2, 1, sharex=False, sharey=False, figsize=(12, 19), height_ratios=[4, 1]
)

train_h3.plot(
    color="orange",
    markersize=0.1,
    ax=axes[0],
    label="train",
    alpha=np.minimum(np.power(train_h3[chicago_crime.target] + 0.4, 2), 1),
)
test_h3.plot(
    color="royalblue",
    markersize=0.1,
    ax=axes[0],
    label="test",
    alpha=np.minimum(np.power(test_h3[chicago_crime.target] + 0.4, 2), 1),
)

cx.add_basemap(axes[0], source=cx.providers.CartoDB.PositronNoLabels, crs=4326, zoom=12)
axes[0].set_title("Chicago crime data aggregated to H3 cells")
axes[0].legend(
    handles=[Patch(facecolor="orange"), Patch(facecolor="royalblue")],
    labels=["Train", "Test"],
)
axes[0].set_axis_off()

sns.kdeplot(
    x=train_h3[chicago_crime.target],
    label="train",
    color="orange",
    ax=axes[1],
    fill=False,
    cut=0,
)
sns.kdeplot(
    x=test_h3[chicago_crime.target],
    label="test",
    color="royalblue",
    ax=axes[1],
    fill=False,
    cut=0,
)
axes[1].set_title("Chicago crime data - target distribution")
axes[1].legend()

fig.tight_layout()

plt.show()

No description has been provided for this image

Load data from 2022

In [13]:

Copied!

ds = chicago_crime.load(version="2022")
ds.keys()
ds = chicago_crime.load(version="2022")
ds.keys()

Out[13]:

dict_keys(['train'])

In [14]:

Copied!

type(chicago_crime.train_gdf), type(chicago_crime.test_gdf)
type(chicago_crime.train_gdf), type(chicago_crime.test_gdf)

Out[14]:

(geopandas.geodataframe.GeoDataFrame, NoneType)

In [15]:

Copied!

ds["train"].head()
ds["train"].head()

Out[15]:

	ID	Case Number	Date	Block	IUCR	Primary Type	Description	Location Description	Arrest	Domestic	Beat	District	Ward	Community Area	FBI Code	Year	Updated On	geometry
0	12589893	JF109865	01/11/2022 03:00:00 PM	087XX S KINGSTON AVE	1565	SEX OFFENSE	INDECENT SOLICITATION OF A CHILD	RESIDENCE	False	True	423	4	7.0	46.0	17	2022	09/14/2023 03:41:59 PM	POINT (-87.56241 41.73641)
1	12592454	JF113025	01/14/2022 03:55:00 PM	067XX S MORGAN ST	2826	OTHER OFFENSE	HARASSMENT BY ELECTRONIC MEANS	RESIDENCE	False	True	724	7	16.0	68.0	26	2022	09/14/2023 03:41:59 PM	POINT (-87.64944 41.77178)
2	12785595	JF346553	08/05/2022 09:00:00 PM	072XX S UNIVERSITY AVE	1544	SEX OFFENSE	SEXUAL EXPLOITATION OF A CHILD	APARTMENT	True	False	324	3	5.0	69.0	17	2022	09/14/2023 03:41:59 PM	POINT (-87.597 41.76334)
3	12808281	JF373517	08/14/2022 02:00:00 PM	055XX W ARDMORE AVE	1562	SEX OFFENSE	AGGRAVATED CRIMINAL SEXUAL ABUSE	RESIDENCE	False	False	1621	16	39.0	11.0	17	2022	09/14/2023 03:41:59 PM	POINT (-87.7664 41.98588)
4	12888104	JF469015	11/10/2022 03:47:00 AM	072XX S MAY ST	1477	WEAPONS VIOLATION	RECKLESS FIREARM DISCHARGE	STREET	False	False	733	7	17.0	68.0	15	2022	09/14/2023 03:41:59 PM	POINT (-87.65284 41.76261)

Create your own train-test split -> Spatial splitting with bucket stratification

In [16]:

Copied!

train, test = chicago_crime.train_test_split(
    test_size=0.2, random_state=42, n_bins=10, resolution=9
)
train, test = chicago_crime.train_test_split(
    test_size=0.2, random_state=42, n_bins=10, resolution=9
)

Summary of the split:

  Train: 3886 H3 cells (187991 points)
  Test: 1075 H3 cells (46928 points)

  Expected ratios: {'train': 0.8, 'validation': 0, 'test': 0.2}
  Actual ratios: {'train': 0.8, 'test': 0.2}
  Actual ratios difference: {'train': 0.0, 'test': 0.0}
   bucket  train_ratio  test_ratio  train_ratio_difference  \
0       0      0.80017     0.19983                -0.00017   
1       1      0.79926     0.20074                 0.00074   
2       2      0.80101     0.19899                -0.00101   
3       3      0.80010     0.19990                -0.00010   
4       4      0.79906     0.20094                 0.00094   
5       5      0.79969     0.20031                 0.00031   
6       6      0.79991     0.20009                 0.00009   
7       7      0.80072     0.19928                -0.00072   
8       8      0.80025     0.19975                -0.00025   
9       9      0.80046     0.19954                -0.00046   

   test_ratio_difference  train_points  test_points  
0                0.00017           965          241  
1               -0.00074          3038          763  
2                0.00101          5531         1374  
3                0.00010          7825         1955  
4               -0.00094         11095         2790  
5               -0.00031         14388         3604  
6               -0.00009         19777         4947  
7                0.00072         25084         6243  
8                0.00025         34607         8638  
9                0.00046         65681        16373  
Created new train_gdf and test_gdf. Train len: 187991,test len: 46928

In [17]:

Copied!

type(chicago_crime.train_gdf), type(chicago_crime.test_gdf)
type(chicago_crime.train_gdf), type(chicago_crime.test_gdf)

Out[17]:

(geopandas.geodataframe.GeoDataFrame, geopandas.geodataframe.GeoDataFrame)

In [18]:

Copied!

chicago_crime.resolution
chicago_crime.resolution

Out[18]:

In [19]:

Copied!

train.head()
train.head()

Out[19]:

	ID	Case Number	Date	Block	IUCR	Primary Type	Description	Location Description	Arrest	Domestic	Beat	District	Ward	Community Area	FBI Code	Year	Updated On	geometry
0	12589893	JF109865	01/11/2022 03:00:00 PM	087XX S KINGSTON AVE	1565	SEX OFFENSE	INDECENT SOLICITATION OF A CHILD	RESIDENCE	False	True	423	4	7.0	46.0	17	2022	09/14/2023 03:41:59 PM	POINT (-87.56241 41.73641)
1	12592454	JF113025	01/14/2022 03:55:00 PM	067XX S MORGAN ST	2826	OTHER OFFENSE	HARASSMENT BY ELECTRONIC MEANS	RESIDENCE	False	True	724	7	16.0	68.0	26	2022	09/14/2023 03:41:59 PM	POINT (-87.64944 41.77178)
2	12785595	JF346553	08/05/2022 09:00:00 PM	072XX S UNIVERSITY AVE	1544	SEX OFFENSE	SEXUAL EXPLOITATION OF A CHILD	APARTMENT	True	False	324	3	5.0	69.0	17	2022	09/14/2023 03:41:59 PM	POINT (-87.597 41.76334)
3	12808281	JF373517	08/14/2022 02:00:00 PM	055XX W ARDMORE AVE	1562	SEX OFFENSE	AGGRAVATED CRIMINAL SEXUAL ABUSE	RESIDENCE	False	False	1621	16	39.0	11.0	17	2022	09/14/2023 03:41:59 PM	POINT (-87.7664 41.98588)
4	12888104	JF469015	11/10/2022 03:47:00 AM	072XX S MAY ST	1477	WEAPONS VIOLATION	RECKLESS FIREARM DISCHARGE	STREET	False	False	733	7	17.0	68.0	15	2022	09/14/2023 03:41:59 PM	POINT (-87.65284 41.76261)

In [20]:

Copied!

test.head()
test.head()

Out[20]:

	ID	Case Number	Date	Block	IUCR	Primary Type	Description	Location Description	Arrest	Domestic	Beat	District	Ward	Community Area	FBI Code	Year	Updated On	geometry
20	12605225	JF128458	01/30/2022 05:30:00 AM	007XX W COUCH PL	0486	BATTERY	DOMESTIC BATTERY SIMPLE	APARTMENT	True	True	1224	12	27.0	28.0	08B	2022	09/16/2023 03:41:56 PM	POINT (-87.64682 41.88511)
22	12622840	JF150396	02/20/2022 10:45:00 AM	010XX W ROSCOE ST	0486	BATTERY	DOMESTIC BATTERY SIMPLE	APARTMENT	True	True	1924	19	44.0	6.0	08B	2022	09/16/2023 03:41:56 PM	POINT (-87.65582 41.94354)
26	12812832	JF378884	06/23/2022 05:35:00 PM	026XX N MOODY AVE	0281	CRIMINAL SEXUAL ASSAULT	NON-AGGRAVATED	APARTMENT	False	True	2512	25	36.0	19.0	02	2022	09/16/2023 03:41:56 PM	POINT (-87.77952 41.92824)
35	12628255	JF157037	02/26/2022 05:45:00 PM	002XX W 24TH PL	0326	ROBBERY	AGGRAVATED VEHICULAR HIJACKING	STREET	False	False	914	9	11.0	34.0	03	2022	10/03/2023 03:41:27 PM	POINT (-87.63357 41.84823)
50	12646302	JF175150	03/11/2022 02:10:00 PM	105XX S MORGAN ST	0266	CRIMINAL SEXUAL ASSAULT	PREDATORY	STREET	False	False	2232	22	21.0	73.0	02	2022	09/29/2023 03:41:29 PM	POINT (-87.64756 41.7024)

In [ ]: