In [1]:

    

import pandas as pd
import numpy as np


    

In [6]:

    

df = pd.read_pickle('../data_processeing/Yelp_Cuisine_Chinese.pkl')
print df.shape
df.head(1)


    

    




(27314, 98)






    
Out[6]:






  

    

      

      
address
      
attributes
      
business_id
      
categories
      
city
      
hours
      
is_open
      
latitude
      
longitude
      
name
      
...
      
RestaurantsDelivery
      
RestaurantsGoodForGroups
      
RestaurantsPriceRange2
      
RestaurantsReservations
      
RestaurantsTableService
      
RestaurantsTakeOut
      
Smoking
      
WheelchairAccessible
      
WiFi
      
cuisine_Chinese
    
  
  

    

      
EDqCEAGXVGCH4FJXgqtjqg
      
979 Bloor Street W
      
[{u'Alcohol': u'none'}, {u'Ambience': {u'roman...
      
EDqCEAGXVGCH4FJXgqtjqg
      
[Restaurants, Pizza, Chicken Wings, Italian]
      
Toronto
      
[Monday 11:0-2:0, Tuesday 11:0-2:0, Wednesday ...
      
1
      
43.661054
      
-79.429089
      
Pizza Pizza
      
...
      
False
      
False
      
NaN
      
NaN
      
NaN
      
NaN
      
False
      
NaN
      
NaN
      
1
    
  

1 rows × 98 columns

In [7]:

    

df.groupby(df.name).size()[df.groupby(df.name).size().values > 100]


    

    
Out[7]:





name
Chipotle Mexican Grill    156
Domino's Pizza            170
KFC                       116
McDonald's                139
Panda Express             123
Panera Bread              119
Pizza Hut                 108
Subway                    160
Taco Bell                 215
dtype: int64

In [8]:

    

df_cn = df[df.cuisine_Chinese == 2]
df_cn.groupby(df_cn.name).size()[df_cn.groupby(df_cn.name).size().values > 10]


    

    
Out[8]:





name
Noodles & Company      19
P.F. Chang's           15
Panda Express         123
Pei Wei                20
Teriyaki Madness       12
The Captain's Boil     11
dtype: int64

In [9]:

    

df2 = df.copy()
df2 = df2[df2.cuisine_Chinese != 0]
print len(df), df2.shape
df2.dropna(axis=1, inplace=True, how='all')
print df2.shape


    

    




27314 (18856, 98)
(18856, 90)

In [11]:

    

df2.groupby(df2.cuisine_Chinese).size()


    

    
Out[11]:





cuisine_Chinese
1    16163
2     2693
dtype: int64

In [6]:

    

df_att = df2.loc[ :, u'AcceptsInsurance':'WiFi'] # include 'WiFi'
att_type = {}
col_b = []
col_m = []
for i in range(df_att.shape[1]):
    temp = df_att.groupby(df_att.iloc[:,i]).size().keys()
    print temp.name, ': ', temp.values
    if set(temp.values) == set([False, True]) or set(temp.values).issubset(set([False, True])):
        att_type[temp.name] = 'b'
        col_b.append(temp.name)
    else:
        att_type[temp.name] = 'm'
        col_m.append(temp.name)

print len(col_b), len(col_m)
col_all = col_b + col_m


    

    




AcceptsInsurance :  [False True]
AgesAllowed :  [False True]
Alcohol :  [False True]
Ambience_casual :  [False True]
Ambience_classy :  ['beer_and_wine' 'full_bar' 'none']
Ambience_divey :  [False True]
Ambience_hipster :  ['average' 'loud' 'quiet' 'very_loud']
Ambience_intimate :  [False True]
Ambience_romantic :  ['casual' 'dressy' 'formal']
Ambience_touristy :  [False True]
Ambience_trendy :  [False True]
Ambience_upscale :  [False True]
BYOB :  [False True]
BYOBCorkage :  [False True]
BestNights_friday :  [False]
BestNights_monday :  [False True]
BestNights_saturday :  [False True]
BestNights_sunday :  [False True]
BestNights_thursday :  [False 'yes_corkage' 'yes_free']
BestNights_tuesday :  [False True]
BestNights_wednesday :  [False True]
BikeParking :  [False True]
BusinessAcceptsBitcoin :  [False True]
BusinessAcceptsCreditCards :  [False True]
BusinessParking_garage :  [False True]
BusinessParking_lot :  [False 'free' 'paid']
BusinessParking_valet :  [False True]
BusinessParking_validated :  [False True]
ByAppointmentOnly :  [False True]
CoatCheck :  [False True]
Corkage :  [False True]
DietaryRestrictions_dairy-free :  [False True]
DietaryRestrictions_halal :  [False True]
DietaryRestrictions_kosher :  [False True]
DietaryRestrictions_soy-free :  [False True]
DietaryRestrictions_vegan :  [False True]
DietaryRestrictions_vegetarian :  [False True]
DogsAllowed :  [False True]
GoodForDancing :  [False True]
GoodForKids :  [False True]
GoodForMeal_breakfast :  [False True]
GoodForMeal_brunch :  [False True]
GoodForMeal_dessert :  [False True]
GoodForMeal_dinner :  [False True]
GoodForMeal_latenight :  [False True]
GoodForMeal_lunch :  [False True]
HairSpecializesIn_africanamerican :  [False True]
HairSpecializesIn_asian :  [False True]
HairSpecializesIn_coloring :  [False True]
HairSpecializesIn_curly :  [False True]
HairSpecializesIn_kids :  [1 2 3 4]
HairSpecializesIn_perms :  ['18plus' '19plus' '21plus' 'allages']
HairSpecializesIn_straightperms :  [False True]
HappyHour :  [False True]
HasTV :  [False True]
Music_background_music :  [False True]
Music_dj :  [False True]
Music_jukebox :  [False True]
Music_live :  [False True]
Music_no_music :  [False True]
Music_video :  [False True]
Open24Hours :  [False True]
OutdoorSeating :  [False True]
RestaurantsAttire :  [False True]
RestaurantsCounterService :  [False True]
RestaurantsDelivery :  [False True]
RestaurantsGoodForGroups :  [False True]
RestaurantsPriceRange2 :  [False True]
RestaurantsReservations :  [False True 'outdoor']
RestaurantsTableService :  [False True]
RestaurantsTakeOut :  [False True]
Smoking :  [False True]
WiFi :  [False True]
65 8

In [7]:

    

df_att_b = df2.loc[:,col_b].join(df.cuisine_Chinese)
df_att_b.head(1).iloc[:,:-1]


    

    
Out[7]:






  

    

      

      
AcceptsInsurance
      
AgesAllowed
      
Alcohol
      
Ambience_casual
      
Ambience_divey
      
Ambience_intimate
      
Ambience_touristy
      
Ambience_trendy
      
Ambience_upscale
      
BYOB
      
...
      
OutdoorSeating
      
RestaurantsAttire
      
RestaurantsCounterService
      
RestaurantsDelivery
      
RestaurantsGoodForGroups
      
RestaurantsPriceRange2
      
RestaurantsTableService
      
RestaurantsTakeOut
      
Smoking
      
WiFi
    
  
  

    

      
EDqCEAGXVGCH4FJXgqtjqg
      
NaN
      
False
      
NaN
      
NaN
      
True
      
NaN
      
NaN
      
NaN
      
True
      
False
      
...
      
False
      
NaN
      
NaN
      
False
      
False
      
NaN
      
NaN
      
NaN
      
False
      
NaN
    
  

1 rows × 65 columns

In [8]:

    

df_att_m = df2.loc[:,col_m].join(df.cuisine_Chinese)
df_att_m.head(1)


    

    
Out[8]:






  

    

      

      
Ambience_classy
      
Ambience_hipster
      
Ambience_romantic
      
BestNights_thursday
      
BusinessParking_lot
      
HairSpecializesIn_kids
      
HairSpecializesIn_perms
      
RestaurantsReservations
      
cuisine_Chinese
    
  
  

    

      
EDqCEAGXVGCH4FJXgqtjqg
      
none
      
quiet
      
casual
      
NaN
      
free
      
1
      
NaN
      
NaN
      
1
    
  

In [9]:

    

df_att_b.to_csv('df_att_b.csv',index_label=False)
df_att_m.to_csv('df_att_m.csv',index_label=False)
df_att_all = df_att.join(df.cuisine_Chinese)
df_att_all.to_csv('df_att_all.csv',index_label=False)


    

In [10]:

    

# from sklearn.naive_bayes import MultinomialNB, BernoulliNB


    

In [11]:

    

# mod_b = BernoulliNB(alpha=1.0, binarize=0.0, fit_prior=True, class_prior=None)


    

In [12]:

    

# mod_b.fit(X=df_att_b.iloc[:,:-1], y=df_att_b.cuisine_Chinese)