In [1]:

    

ls


    

    




sampleSubmission.csv  starter.ipynb  test.csv  train.csv

In [2]:

    

import pandas as pd


    

    




Vendor:  Continuum Analytics, Inc.
Package: mkl
Message: trial mode expires in 16 days
Vendor:  Continuum Analytics, Inc.
Package: mkl
Message: trial mode expires in 16 days

In [3]:

    

train_frame = pd.read_csv('train.csv')


    

In [4]:

    

train_frame.columns


    

    
Out[4]:





Index([u'Dates', u'Category', u'Descript', u'DayOfWeek', u'PdDistrict', u'Resolution', u'Address', u'X', u'Y'], dtype='object')

In [61]:

    

d = dict()
for category, table in train_frame.groupby(by="Category"):
    d.update({category:table['Descript'].unique()})


    

In [ ]:

    




    

In [63]:

    

d.keys()


    

    
Out[63]:





['RECOVERED VEHICLE',
 'SUICIDE',
 'FRAUD',
 'WEAPON LAWS',
 'VANDALISM',
 'ARSON',
 'OTHER OFFENSES',
 'WARRANTS',
 'LOITERING',
 'DRUG/NARCOTIC',
 'EMBEZZLEMENT',
 'SEX OFFENSES NON FORCIBLE',
 'KIDNAPPING',
 'DRIVING UNDER THE INFLUENCE',
 'LARCENY/THEFT',
 'ROBBERY',
 'MISSING PERSON',
 'BURGLARY',
 'RUNAWAY',
 'STOLEN PROPERTY',
 'PORNOGRAPHY/OBSCENE MAT',
 'SUSPICIOUS OCC',
 'DISORDERLY CONDUCT',
 'LIQUOR LAWS',
 'FAMILY OFFENSES',
 'TRESPASS',
 'TREA',
 'SECONDARY CODES',
 'VEHICLE THEFT',
 'BAD CHECKS',
 'SEX OFFENSES FORCIBLE',
 'FORGERY/COUNTERFEITING',
 'ASSAULT',
 'BRIBERY',
 'NON-CRIMINAL',
 'GAMBLING',
 'EXTORTION',
 'PROSTITUTION',
 'DRUNKENNESS']

In [66]:

    

d['KIDNAPPING']


    

    
Out[66]:





array(['FALSE IMPRISONMENT', 'KIDNAPPING DURING ROBBERY', 'CHILD STEALING',
       'KIDNAPPING, ADULT VICTIM', 'ATTEMPTED KIDNAPPING, JUVENILE VICTIM',
       'KIDNAPPING, JUVENILE VICTIM', 'KIDNAPPING DURING CARJACKING',
       'ATTEMPTED KIDNAPPING, ADULT VICTIM', 'KIDNAPPER, POSING AS'], dtype=object)

In [25]:

    

train_frame.Category.describe()


    

    
Out[25]:





count            878049
unique               39
top       LARCENY/THEFT
freq             174900
Name: Category, dtype: object

In [ ]:

    




    

In [29]:

    

train_frame.groupby(by=train_frame.Category).describe()


    

    
Out[29]:






  

    

      

      

      
X
      
Y
    
    

      
Category
      

      

      

    
  
  

    

      
ARSON
      
count
      
  1513.000000
      
  1513.000000
    
    

      
mean
      
  -122.419799
      
    37.757478
    
    

      
std
      
     0.029780
      
     0.027173
    
    

      
min
      
  -122.510037
      
    37.708154
    
    

      
25%
      
  -122.433892
      
    37.732303
    
    

      
50%
      
  -122.414544
      
    37.761090
    
    

      
75%
      
  -122.399129
      
    37.780978
    
    

      
max
      
  -122.364937
      
    37.819923
    
    

      
ASSAULT
      
count
      
 76876.000000
      
 76876.000000
    
    

      
mean
      
  -122.421062
      
    37.766595
    
    

      
std
      
     0.028574
      
     0.377655
    
    

      
min
      
  -122.513642
      
    37.707922
    
    

      
25%
      
  -122.430759
      
    37.743555
    
    

      
50%
      
  -122.415722
      
    37.772541
    
    

      
75%
      
  -122.406670
      
    37.783672
    
    

      
max
      
  -120.500000
      
    90.000000
    
    

      
BAD CHECKS
      
count
      
   406.000000
      
   406.000000
    
    

      
mean
      
  -122.423752
      
    37.769944
    
    

      
std
      
     0.024956
      
     0.023670
    
    

      
min
      
  -122.506213
      
    37.708816
    
    

      
25%
      
  -122.437571
      
    37.752700
    
    

      
50%
      
  -122.416197
      
    37.778263
    
    

      
75%
      
  -122.405176
      
    37.787689
    
    

      
max
      
  -122.365565
      
    37.809671
    
    

      
BRIBERY
      
count
      
   289.000000
      
   289.000000
    
    

      
mean
      
  -122.418650
      
    37.754059
    
    

      
std
      
     0.025421
      
     0.024816
    
    

      
min
      
  -122.505928
      
    37.709030
    
    

      
25%
      
  -122.430701
      
    37.731740
    
    

      
50%
      
  -122.416078
      
    37.754626
    
    

      
...
      
...
      
...
      
...
    
    

      
VANDALISM
      
std
      
     0.030115
      
     0.248353
    
    

      
min
      
  -122.513642
      
    37.707922
    
    

      
25%
      
  -122.439631
      
    37.742772
    
    

      
50%
      
  -122.419483
      
    37.769955
    
    

      
75%
      
  -122.406691
      
    37.783310
    
    

      
max
      
  -120.500000
      
    90.000000
    
    

      
VEHICLE THEFT
      
count
      
 53781.000000
      
 53781.000000
    
    

      
mean
      
  -122.429060
      
    37.768329
    
    

      
std
      
     0.037893
      
     0.676246
    
    

      
min
      
  -122.513642
      
    37.707920
    
    

      
25%
      
  -122.443399
      
    37.737311
    
    

      
50%
      
  -122.423602
      
    37.762628
    
    

      
75%
      
  -122.409530
      
    37.780849
    
    

      
max
      
  -120.500000
      
    90.000000
    
    

      
WARRANTS
      
count
      
 42214.000000
      
 42214.000000
    
    

      
mean
      
  -122.417581
      
    37.778991
    
    

      
std
      
     0.033304
      
     0.719289
    
    

      
min
      
  -122.513642
      
    37.707922
    
    

      
25%
      
  -122.421305
      
    37.760957
    
    

      
50%
      
  -122.414056
      
    37.776231
    
    

      
75%
      
  -122.407474
      
    37.783730
    
    

      
max
      
  -120.500000
      
    90.000000
    
    

      
WEAPON LAWS
      
count
      
  8555.000000
      
  8555.000000
    
    

      
mean
      
  -122.418983
      
    37.758734
    
    

      
std
      
     0.025308
      
     0.026170
    
    

      
min
      
  -122.510226
      
    37.707922
    
    

      
25%
      
  -122.427918
      
    37.734491
    
    

      
50%
      
  -122.414810
      
    37.764057
    
    

      
75%
      
  -122.405284
      
    37.781605
    
    

      
max
      
  -122.365565
      
    37.819923
    
  

312 rows × 2 columns

In [ ]:

    




    

In [9]:

    

categories.


    

    




---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-9-8de2132cadb8> in <module>()
----> 1 categories.columns

/home/will/anaconda/lib/python2.7/site-packages/pandas/core/groupby.pyc in __getattr__(self, attr)
    506             return self[attr]
    507         if hasattr(self.obj, attr):
--> 508             return self._make_wrapper(attr)
    509 
    510         raise AttributeError("%r object has no attribute %r" %

/home/will/anaconda/lib/python2.7/site-packages/pandas/core/groupby.pyc in _make_wrapper(self, name)
    521                    "using the 'apply' method".format(kind, name,
    522                                                      type(self).__name__))
--> 523             raise AttributeError(msg)
    524 
    525         # need to setup the selection

AttributeError: Cannot access attribute 'columns' of 'DataFrameGroupBy' objects, try using the 'apply' method

In [ ]: