In [1]:

    

%matplotlib inline
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.neighbors import KernelDensity


    

In [2]:

    

df_train = pd.read_csv("/Users/dikien/Downloads/West Nile Virus Prediction/train.csv")
df_weather = pd.read_csv("/Users/dikien/Downloads/West Nile Virus Prediction/weather.csv")
df_spray = pd.read_csv("/Users/dikien/Downloads/West Nile Virus Prediction/spray.csv")


    

In [3]:

    

print df_train.head(2)
print df_train.describe() 
print "=" * 90
print df_weather.head(2)
print df_weather.describe()
print "=" * 90
print df_spray.head(2)
print df_spray.describe()


    

    




         Date                                            Address  \
0  2007-05-29  4100 North Oak Park Avenue, Chicago, IL 60634,...   
1  2007-05-29  4100 North Oak Park Avenue, Chicago, IL 60634,...   

                  Species  Block           Street  Trap  \
0  CULEX PIPIENS/RESTUANS     41   N OAK PARK AVE  T002   
1          CULEX RESTUANS     41   N OAK PARK AVE  T002   

              AddressNumberAndStreet  Latitude  Longitude  AddressAccuracy  \
0  4100  N OAK PARK AVE, Chicago, IL  41.95469 -87.800991                9   
1  4100  N OAK PARK AVE, Chicago, IL  41.95469 -87.800991                9   

   NumMosquitos  WnvPresent  
0             1           0  
1             1           0  
              Block      Latitude     Longitude  AddressAccuracy  \
count  10506.000000  10506.000000  10506.000000     10506.000000   
mean      35.687797     41.841139    -87.699908         7.819532   
std       24.339468      0.112742      0.096514         1.452921   
min       10.000000     41.644612    -87.930995         3.000000   
25%       12.000000     41.732984    -87.760070         8.000000   
50%       33.000000     41.846283    -87.694991         8.000000   
75%       52.000000     41.954690    -87.627796         9.000000   
max       98.000000     42.017430    -87.531635         9.000000   

       NumMosquitos    WnvPresent  
count  10506.000000  10506.000000  
mean      12.853512      0.052446  
std       16.133816      0.222936  
min        1.000000      0.000000  
25%        2.000000      0.000000  
50%        5.000000      0.000000  
75%       17.000000      0.000000  
max       50.000000      1.000000  
==========================================================================================
   Station        Date  Tmax  Tmin Tavg Depart  DewPoint WetBulb Heat Cool  \
0        1  2007-05-01    83    50   67     14        51      56    0    2   
1        2  2007-05-01    84    52   68      M        51      57    0    3   

     ...    CodeSum Depth Water1 SnowFall PrecipTotal StnPressure SeaLevel  \
0    ...                0      M      0.0        0.00       29.10    29.82   
1    ...                M      M        M        0.00       29.18    29.82   

  ResultSpeed ResultDir  AvgSpeed  
0         1.7        27       9.2  
1         2.7        25       9.6  

[2 rows x 22 columns]
           Station         Tmax         Tmin     DewPoint  ResultSpeed  \
count  2944.000000  2944.000000  2944.000000  2944.000000  2944.000000   
mean      1.500000    76.166101    57.810462    53.457880     6.960666   
std       0.500085    11.461970    10.381939    10.675181     3.587527   
min       1.000000    41.000000    29.000000    22.000000     0.100000   
25%       1.000000    69.000000    50.000000    46.000000     4.300000   
50%       1.500000    78.000000    59.000000    54.000000     6.400000   
75%       2.000000    85.000000    66.000000    62.000000     9.200000   
max       2.000000   104.000000    83.000000    75.000000    24.100000   

         ResultDir  
count  2944.000000  
mean     17.494905  
std      10.063609  
min       1.000000  
25%       7.000000  
50%      19.000000  
75%      25.000000  
max      36.000000  
==========================================================================================
         Date        Time   Latitude  Longitude
0  2011-08-29  6:56:58 PM  42.391623 -88.089163
1  2011-08-29  6:57:08 PM  42.391348 -88.089163
           Latitude     Longitude
count  14835.000000  14835.000000
mean      41.904828    -87.736690
std        0.104381      0.067292
min       41.713925    -88.096468
25%       41.785001    -87.794225
50%       41.940075    -87.727853
75%       41.980978    -87.694108
max       42.395983    -87.586727

In [5]:

    

mapdata = np.loadtxt("/Users/dikien/Downloads/West Nile Virus Prediction/mapdata_copyright_openstreetmap_contributors.txt")
traps = pd.read_csv("/Users/dikien/Downloads/West Nile Virus Prediction/train.csv")[['Date', 'Trap','Longitude', 'Latitude', 'WnvPresent']]


    

In [7]:

    

alpha_cm = plt.cm.Reds
alpha_cm._init()
alpha_cm._lut[:-3,-1] = abs(np.logspace(0, 1, alpha_cm.N) / 10 - 1)[::-1]
aspect = mapdata.shape[0] * 1.0 / mapdata.shape[1]
lon_lat_box = (-88, -87.5, 41.6, 42.1)


    

In [8]:

    

sigthings = traps[traps['WnvPresent'] > 0]
sigthings = sigthings.groupby(['Date', 'Trap','Longitude', 'Latitude']).max()['WnvPresent'].reset_index()
X = sigthings[['Longitude', 'Latitude']].values
kd = KernelDensity(bandwidth=0.02)
kd.fit(X)


    

    
Out[8]:





KernelDensity(algorithm='auto', atol=0, bandwidth=0.02, breadth_first=True,
       kernel='gaussian', leaf_size=40, metric='euclidean',
       metric_params=None, rtol=0)

In [9]:

    

xv,yv = np.meshgrid(np.linspace(-88, -87.5, 100), np.linspace(41.6, 42.1, 100))
gridpoints = np.array([xv.ravel(),yv.ravel()]).T
zv = np.exp(kd.score_samples(gridpoints).reshape(100,100))


    

In [10]:

    

plt.figure(figsize=(10,14))
plt.imshow(mapdata, 
           cmap=plt.get_cmap('gray'), 
           extent=lon_lat_box, 
           aspect=aspect)
plt.imshow(zv, 
           origin='lower', 
           cmap=alpha_cm, 
           extent=lon_lat_box, 
           aspect=aspect)

locations = traps[['Longitude', 'Latitude']].drop_duplicates().values
plt.scatter(locations[:,0], locations[:,1], marker='x')


    

    
Out[10]:





<matplotlib.collections.PathCollection at 0x109c07410>






    

In [11]:

    

df_test = pd.read_csv("/Users/dikien/Downloads/West Nile Virus Prediction/test.csv")


    

In [12]:

    

df_test.columns


    

    
Out[12]:





Index([u'Id', u'Date', u'Address', u'Species', u'Block', u'Street', u'Trap',
       u'AddressNumberAndStreet', u'Latitude', u'Longitude',
       u'AddressAccuracy'],
      dtype='object')

In [17]:

    

df = pd.read_csv("/Users/dikien/Downloads/West Nile Virus Prediction/sampleSubmission.csv", dtype={'Id': np.int, 'WnvPresent' : np.float32})


    

In [19]:

    

df.dtypes


    

    
Out[19]:





Id              int64
WnvPresent    float32
dtype: object