In [1]:

    

import holoviews as hv
import geoviews as gv
import param, paramnb, parambokeh
import pandas as pd
import dask.dataframe as dd
import datashader as ds

from colorcet import cm
from bokeh.models import WMTSTileSource
from holoviews.operation.datashader import datashade
from holoviews.operation import decimate
from holoviews.streams import RangeXY, PlotSize


    

    




D:\tools\dev\python\Anaconda3-4.2.0\lib\site-packages\seaborn\apionly.py:6: UserWarning: As seaborn no longer sets a default style on import, the seaborn.apionly module is deprecated. It will be removed in a future version.
  warnings.warn(msg, UserWarning)
D:\tools\dev\python\Anaconda3-4.2.0\lib\site-packages\odo\backends\pandas.py:94: FutureWarning: pandas.tslib is deprecated and will be removed in a future version.
You can access NaTType as type(pandas.NaT)
  @convert.register((pd.Timestamp, pd.Timedelta), (pd.tslib.NaTType, type(None)))

In [2]:

    

hv.extension('bokeh')


    

In [3]:

    

%%time
df = dd.read_parquet('../data/crimes-2017.snappy.parq/').persist()
print(len(df))


    

    




146711
Wall time: 2.71 s

In [4]:

    

df.head()


    

    
Out[4]:







  

    

      

      
Date
      
Block
      
PrimaryType
      
Description
      
LocationDescription
      
Arrest
      
Domestic
      
Latitude
      
Longitude
    
  
  

    

      
0
      
2017-07-22 04:39:00
      
082XX S EMERALD AVE
      
HOMICIDE
      
FIRST DEGREE MURDER
      
STREET
      
False
      
False
      
NaN
      
NaN
    
    

      
1
      
2017-07-25 22:32:00
      
078XX S CARPENTER ST
      
HOMICIDE
      
FIRST DEGREE MURDER
      
STREET
      
False
      
False
      
41.751545
      
-87.650128
    
    

      
2
      
2017-07-25 06:25:00
      
003XX N LOREL AVE
      
HOMICIDE
      
FIRST DEGREE MURDER
      
ALLEY
      
False
      
False
      
41.886330
      
-87.758956
    
    

      
3
      
2017-02-15 11:20:00
      
003XX S CANAL ST
      
HOMICIDE
      
FIRST DEGREE MURDER
      
STREET
      
True
      
False
      
NaN
      
NaN
    
    

      
4
      
2017-07-14 09:00:00
      
014XX W 112TH PL
      
BURGLARY
      
FORCIBLE ENTRY
      
VACANT LOT/LAND
      
False
      
False
      
NaN
      
NaN
    
  

In [5]:

    

# drop records with NaN coordinates
df = df.dropna(how='any', subset=['Latitude', 'Longitude']) 
df.head()


    

    
Out[5]:







  

    

      

      
Date
      
Block
      
PrimaryType
      
Description
      
LocationDescription
      
Arrest
      
Domestic
      
Latitude
      
Longitude
    
  
  

    

      
1
      
2017-07-25 22:32:00
      
078XX S CARPENTER ST
      
HOMICIDE
      
FIRST DEGREE MURDER
      
STREET
      
False
      
False
      
41.751545
      
-87.650128
    
    

      
2
      
2017-07-25 06:25:00
      
003XX N LOREL AVE
      
HOMICIDE
      
FIRST DEGREE MURDER
      
ALLEY
      
False
      
False
      
41.886330
      
-87.758956
    
    

      
59
      
2017-07-25 11:15:00
      
070XX S YALE AVE
      
MOTOR VEHICLE THEFT
      
AUTOMOBILE
      
STREET
      
False
      
False
      
41.766408
      
-87.631027
    
    

      
60
      
2017-07-25 00:01:00
      
081XX S CARPENTER ST
      
THEFT
      
$500 AND UNDER
      
STREET
      
False
      
False
      
41.746417
      
-87.649988
    
    

      
61
      
2017-07-25 22:31:00
      
027XX N LINCOLN AVE
      
CRIMINAL DAMAGE
      
TO VEHICLE
      
PARKING LOT/GARAGE(NON.RESID.)
      
False
      
False
      
41.931287
      
-87.656934
    
  

In [6]:

    

# plot coordinates
points = hv.Points(df, kdims=['Longitude', 'Latitude'], vdims=['PrimaryType'], label='Map')
decimate(points) # limit to 1K points max


    

    
Out[6]:






    

        

    

In [7]:

    

# create map tiles
tiles = gv.WMTS(WMTSTileSource(url='https://server.arcgisonline.com/ArcGIS/rest/services/'
                                   'World_Imagery/MapServer/tile/{Z}/{Y}/{X}.jpg'))


    

In [8]:

    

options = dict(width=400,height=400,xaxis=None,yaxis=None,bgcolor='black',show_grid=False)


    

In [ ]:

    

#TODO
gv_dataset = gv.Dataset(df, kdims=['Longitude', 'Latitude'], vdims=['PrimaryType'])
tiles * datashade(gv_dataset, 
                  x_sampling=10, y_sampling=10, 
                  cmap=cm['fire'], 
                  element_type=gv.Image, aggregator=ds.count_cat('PrimaryType')).opts(plot=options)


    

In [ ]:

    

crime_points = datashade(points, 
                         x_sampling=1, y_sampling=1, 
                         cmap=cm['fire'], element_type=gv.Image).opts(plot=options)
crime_points


    

In [ ]: