In [11]:

    

import json, re
import pandas as pd
from datetime import datetime
import numpy as np
import math
from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot
import plotly.graph_objs as go
from plotly import tools
import plotly.plotly as py
import cufflinks as cf
cf.go_online()

inpExifFl = "../data/Flickr_Beauty_features.json"

with open(inpExifFl,"r") as inpJsonFl:
    exifJsonObj = json.load(inpJsonFl)


    

In [16]:

    

exifDf = pd.DataFrame(exifJsonObj).transpose()

# exifDf['datetime'] = exifDf['datetime'].apply(lambda x : datetime.strptime(x,'%Y-%m-%d %H:%M:%S'))
# exifDf['day'] = exifDf.datetime.apply(lambda x : x.day)
# exifDf['hour'] = exifDf.datetime.apply(lambda x : x.hour)
# exifDf.drop(['datetime'],1,inplace=True)
exifDf.head()


    

    
Out[16]:






  

    

      

      
arousal
      
contrast
      
dominance
      
hsv_itten_std_h
      
hsv_itten_std_s
      
hsv_itten_std_v
      
pleasure
      
symmetry
    
  
  

    

      
1
      
-0.105433
      
1.70422
      
0.506566
      
33183.8
      
66438.4
      
32484.3
      
0.45014
      
8.38616
    
    

      
10
      
-0.00690542
      
1.68977
      
0.471749
      
65048.1
      
67045.8
      
71687.8
      
0.410388
      
5.88898
    
    

      
100
      
0.0587454
      
3.70982
      
0.29646
      
12937.7
      
42971.5
      
64439
      
0.251809
      
13.2125
    
    

      
1000
      
0.10483
      
2.33853
      
0.434643
      
36115.3
      
29319.6
      
74245.7
      
0.367374
      
9.01152
    
    

      
1001
      
-0.136607
      
1.27727
      
0.670583
      
22162
      
40191.3
      
55682.2
      
0.595601
      
10.0584
    
  

In [18]:

    

# trace1 = go.Histogram(
#         x=exifDf['height']
#     )

# trace2 = go.Histogram(
#         x=exifDf['width']
#     )

# trace3 = go.Histogram(
#         x=exifDf['lat']
#     )

# trace4 = go.Histogram(
#         x=exifDf['long']
#     )

# trace5 = go.Histogram(
#         x=exifDf['orientation']
#     )

# trace6 = go.Histogram(
#         x=exifDf['day']
#     )

# trace7 = go.Histogram(
#         x=exifDf['hour']
#     )

trace8 = go.Histogram(
        x=exifDf['pleasure']
    )

trace9 = go.Histogram(
        x=exifDf['arousal']
    )

trace10 = go.Histogram(
        x=exifDf['dominance']
    )

trace11 = go.Histogram(
        x=exifDf['contrast']
    )

trace12 = go.Histogram(
        x=exifDf['hsv_itten_std_h']
    )

trace13 = go.Histogram(
        x=exifDf['hsv_itten_std_s']
    )

trace14 = go.Histogram(
        x=exifDf['hsv_itten_std_v']
    )

trace15 = go.Histogram(
        x=exifDf['symmetry']
    )


fig = tools.make_subplots(rows=4, cols=2, subplot_titles=('pleasure','arousal','dominance','contrast','hsv_itten_std_h','hsv_itten_std_s','hsv_itten_std_v','symmetry'))

# fig.append_trace(trace1, 1, 1)
# fig.append_trace(trace2, 1, 2)
# fig.append_trace(trace3, 2, 1)
# fig.append_trace(trace4, 2, 2)
# fig.append_trace(trace5, 3, 1)
# fig.append_trace(trace6, 3, 2)
# fig.append_trace(trace7, 4, 1)
fig.append_trace(trace8, 1, 1)
fig.append_trace(trace9, 1, 2)
fig.append_trace(trace10, 2, 1)
fig.append_trace(trace11, 2, 2)
fig.append_trace(trace12, 3, 1)
fig.append_trace(trace13, 3, 2)
fig.append_trace(trace14, 4, 1)
fig.append_trace(trace15, 4, 2)


fig['layout'].update(showlegend=False, height=1000, width=1000, title='Data distributions')
plot(fig, filename='/tmp/Distributions_BeautyFeatures_Flickr.html')


    

    




This is the format of your plot grid:
[ (1,1) x1,y1 ]  [ (1,2) x2,y2 ]
[ (2,1) x3,y3 ]  [ (2,2) x4,y4 ]
[ (3,1) x5,y5 ]  [ (3,2) x6,y6 ]
[ (4,1) x7,y7 ]  [ (4,2) x8,y8 ]







    
Out[18]:





'file:///tmp/Distributions_BeautyFeatures_Flickr.html'

In [ ]:

    

from numpy import arange,array,ones
from scipy import stats


    

In [37]:

    

resultsDf = pd.DataFrame.from_csv("../FinalResults/resultsExpt2RankList_Tags.csv")

resultsDf = pd.DataFrame(resultsDf['Proportion'])
resultsDict = resultsDf.to_dict(orient="index")

expt2Results = {}

for gid in resultsDict:
    expt2Results[str(gid)] = exifJsonObj[str(gid)]
    expt2Results[str(gid)].update(resultsDict[gid])


    

In [38]:

    

expt2ResultsDf = pd.DataFrame(expt2Results).transpose()

expt2ResultsDf['datetime'] = expt2ResultsDf['datetime'].apply(lambda x : datetime.strptime(x,'%Y-%m-%d %H:%M:%S'))
expt2ResultsDf['day'] = expt2ResultsDf.datetime.apply(lambda x : x.day)
expt2ResultsDf['hour'] = expt2ResultsDf.datetime.apply(lambda x : x.hour)
expt2ResultsDf.drop(['size','datetime'],1,inplace=True)
expt2ResultsDf.head()


    

    
Out[38]:






  

    

      

      
Proportion
      
arousal
      
contrast
      
dominance
      
height
      
hsv_itten_std_h
      
hsv_itten_std_s
      
hsv_itten_std_v
      
lat
      
long
      
orientation
      
pleasure
      
symmetry
      
width
      
day
      
hour
    
  
  

    

      
10
      
20
      
-0.0677241
      
1.4738
      
0.500873
      
4000
      
105719
      
140171
      
133193
      
-1.36709
      
36.782
      
1
      
0.441554
      
11.4251
      
6000
      
1
      
15
    
    

      
1000
      
90
      
0.0267916
      
2.2988
      
0.446958
      
3072
      
100419
      
106270
      
146868
      
-1.34968
      
36.806
      
1
      
0.385603
      
10.8675
      
4608
      
1
      
17
    
    

      
1003
      
20
      
0.00380291
      
1.61078
      
0.534058
      
3072
      
116605
      
145911
      
182645
      
-1.34965
      
36.8062
      
1
      
0.463471
      
9.66617
      
4608
      
1
      
17
    
    

      
1005
      
80
      
-0.049412
      
1.75693
      
0.55702
      
3072
      
102427
      
140823
      
132426
      
-1.34965
      
36.8062
      
1
      
0.488551
      
12.6299
      
4608
      
1
      
17
    
    

      
101
      
20
      
0.0550115
      
2.42374
      
0.444083
      
4000
      
142995
      
193383
      
250416
      
-1.37325
      
36.8003
      
1
      
0.380382
      
6.21572
      
6000
      
1
      
16
    
  

In [ ]:

    

def linearRegressPlot(resultDf,xFtr,yStr):
    xi = np.array(resultDf[xFtr])
    y = list(resultDf[yStr])
    slope, intercept, r_value, p_value, std_err = stats.linregress(xi,y)
    line = slope*xi+intercept
    layout= go.Layout(
            title= "Linear line fit for regression using %s" %xFtr,
            showlegend=False,
            xaxis= dict(
                title= xFtr,
                ticklen= 5,
                zeroline= True,
                gridwidth= 2
            ),
            yaxis=dict(
                title= yStr,
                ticklen= 5,
                gridwidth= 2,
                #range=range
            )
        )

    trace1 = go.Scatter(
                x = xi,
                y = y,
                mode = 'markers'
        )

    trace2 = go.Scatter(
                x=xi, 
                y=line, 
                mode='lines',
                name='Fit'
                      )

    data = [trace1,trace2]

    fig = dict(data=data,layout=layout)
    figMain = py.iplot(fig,filename= str("linear_fit_%s" %xFtr))
    print("r_value, p_value" ,(r_value,p_value))
    print(figMain.embed_code)

linearRegressPlot(expt2ResultsDf,'symmetry','Proportion')


    

In [ ]:

    

xi = np.array(expt2ResultsDf['height'])
y = list(expt2ResultsDf['Proportion'])
slope, intercept, r_value, p_value, std_err = stats.linregress(xi,y)
slope, intercept, r_value, p_value, std_err


    

In [ ]:

    

exifDf.max()


    

In [ ]:

    




    

In [ ]:

    

import importlib

import ExtractBtyFtrs as EBty
importlib.reload(EBty)


    

In [ ]:

    

EBty.createFtrFile("../FinalResults/resultsExpt2RankList_Tags.csv","../data/GZC_exifs_beauty_full.json","../data/BeautyFtrVector_GZC.csv")


    

In [ ]:

    

resultsDf = pd.DataFrame.from_csv("../FinalResults/resultsExpt2RankList_Tags.csv")
resultsDf.head()


    

In [ ]:

    




    

In [67]:

    

from sklearn.model_selection import train_test_split
import ClassiferHelperAPI as CH
import RegressionCapsuleClass as RgrCls
import pandas as pd
import importlib
importlib.reload(RgrCls)

from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot
import plotly.graph_objs as go
import cufflinks as cf
cf.go_offline()


    

In [2]:

    

inpData = pd.DataFrame.from_csv("../data/BeautyFtrVector_GZC.csv")
inpData.head()


    

    
Out[2]:






  

    

      

      
Proportion
      
arousal
      
contrast
      
dominance
      
height
      
hsv_itten_std_h
      
hsv_itten_std_s
      
hsv_itten_std_v
      
lat
      
long
      
orientation
      
pleasure
      
symmetry
      
width
      
day
      
hour
    
  
  

    

      
10
      
20.0
      
-0.067724
      
1.473800
      
0.500873
      
4000
      
105718.831503
      
140170.630774
      
133192.928522
      
-1.367088
      
36.781978
      
1
      
0.441554
      
11.425084
      
6000
      
1
      
15
    
    

      
1000
      
90.0
      
0.026792
      
2.298800
      
0.446958
      
3072
      
100418.690900
      
106270.090353
      
146867.861046
      
-1.349683
      
36.806025
      
1
      
0.385603
      
10.867455
      
4608
      
1
      
17
    
    

      
1003
      
20.0
      
0.003803
      
1.610780
      
0.534058
      
3072
      
116605.125589
      
145911.415673
      
182644.571576
      
-1.349651
      
36.806162
      
1
      
0.463471
      
9.666165
      
4608
      
1
      
17
    
    

      
1005
      
80.0
      
-0.049412
      
1.756933
      
0.557020
      
3072
      
102427.151569
      
140822.729057
      
132426.326397
      
-1.349651
      
36.806162
      
1
      
0.488551
      
12.629949
      
4608
      
1
      
17
    
    

      
101
      
20.0
      
0.055012
      
2.423743
      
0.444083
      
4000
      
142994.891693
      
193382.957863
      
250415.909007
      
-1.373255
      
36.800266
      
1
      
0.380382
      
6.215719
      
6000
      
1
      
16
    
  

In [ ]:

    

inpData = inpData[(inpData['Proportion'] >= 80.0) | (inpData['Proportion'] <= 20.0)]
inpData['TARGET'] = np.where(inpData['Proportion'] >= 80.0, 1, 0)
inpData.drop(['Proportion'],1,inplace=True)


    

In [3]:

    

y = inpData['Proportion']
inpData.drop(['Proportion'],1,inplace=True)


    

In [1]:

    




    

In [17]:

    

train_x, test_x, train_y, test_y = CH.train_test_split(inpData, y, test_size = 0.2)
rgr = CH.getLearningAlgo('elastic_net',{'fit_intercept':True})
    
rgrObj = RgrCls.RegressionCapsule(rgr,'elastic_net',0.8,train_x,train_y,test_x,test_y)


    

In [18]:

    

rgrObj.runRgr(removeOutliers=False)


    

In [19]:

    

rgrObj.abserr, rgrObj.sqerr


    

    
Out[19]:





(23.611381944920133, 796.88386165148518)

In [20]:

    

x_data = rgrObj.preds
y_data = rgrObj.residues


    

In [24]:

    

# Create a trace
trace = go.Scatter(
    x = x_data,
    y = y_data,
    mode = 'markers'
)

data = [trace]

# Plot and embed in ipython notebook!
a = py.iplot(data, filename='basic-scatter')


    

In [26]:

    

a.embed_code


    

    
Out[26]:





'<iframe id="igraph" scrolling="no" style="border:none;" seamless="seamless" src="https://plot.ly/~kamblenoopur/103.embed" height="525px" width="100%"></iframe>'

In [ ]:

    

rgrObj.fit(train_x, train_y)


    

In [ ]:

    

from sklearn.metrics import mean_absolute_error, mean_squared_error


    

In [ ]:

    

mean_squared_error(test_y, preds), mean_absolute_error(test_y, preds)


    

In [ ]:

    

mean_squared_error(test_y, preds), mean_absolute_error(test_y, preds)


    

In [7]:

    

inpData.head()


    

    
Out[7]:






  

    

      

      
arousal
      
contrast
      
dominance
      
height
      
hsv_itten_std_h
      
hsv_itten_std_s
      
hsv_itten_std_v
      
lat
      
long
      
orientation
      
pleasure
      
symmetry
      
width
      
day
      
hour
    
  
  

    

      
10
      
-0.067724
      
1.473800
      
0.500873
      
4000
      
105718.831503
      
140170.630774
      
133192.928522
      
-1.367088
      
36.781978
      
1
      
0.441554
      
11.425084
      
6000
      
1
      
15
    
    

      
1000
      
0.026792
      
2.298800
      
0.446958
      
3072
      
100418.690900
      
106270.090353
      
146867.861046
      
-1.349683
      
36.806025
      
1
      
0.385603
      
10.867455
      
4608
      
1
      
17
    
    

      
1003
      
0.003803
      
1.610780
      
0.534058
      
3072
      
116605.125589
      
145911.415673
      
182644.571576
      
-1.349651
      
36.806162
      
1
      
0.463471
      
9.666165
      
4608
      
1
      
17
    
    

      
1005
      
-0.049412
      
1.756933
      
0.557020
      
3072
      
102427.151569
      
140822.729057
      
132426.326397
      
-1.349651
      
36.806162
      
1
      
0.488551
      
12.629949
      
4608
      
1
      
17
    
    

      
101
      
0.055012
      
2.423743
      
0.444083
      
4000
      
142994.891693
      
193382.957863
      
250415.909007
      
-1.373255
      
36.800266
      
1
      
0.380382
      
6.215719
      
6000
      
1
      
16
    
  

In [1]:

    

import json
import pandas as pd
from datetime import datetime


    

In [12]:

    

with open("../data/GZC_exifs_beauty_full.json") as jsonObj:
    jsonDict = json.load(jsonObj)
df = pd.DataFrame.from_dict(jsonDict).transpose()
# df['datetime'] = df['datetime'].apply(lambda x : datetime.strptime(x,'%Y-%m-%d %H:%M:%S'))
# df['day'] = df.datetime.apply(lambda x : x.day)
# df['hour'] = df.datetime.apply(lambda x : x.hour)
# df.drop(['size','datetime'],1,inplace=True)
# df.head()
# df.to_csv("../data/GZC_exifs_beauty_full.csv")


    

In [5]:

    

len(df)


    

    
Out[5]:





9406

In [161]:

    

import PopulationEstimatorFromClf as PE
import importlib
importlib.reload(PE)

regrArgs = {'linear' : {'fit_intercept' : True},
            'ridge' : {'fit_intercept' : True},
            'lasso' : {'fit_intercept' : True},
            'svr' : {'fit_intercept' : True},
            'dtree_regressor' : {'fit_intercept' : True}}

methArgs = {'dummy' : {'strategy' : 'most_frequent'},
            'bayesian' : {'fit_prior' : True},
            'logistic' : {'penalty' : 'l2'},
            'svm' : {'kernel' : 'rbf','probability' : True},
            'dtree' : {'criterion' : 'entropy'},
            'random_forests' : {'n_estimators' : 10 },
            'ada_boost' : {'n_estimators' : 50 }}

inExifFl,inGidAidMapFl,inAidFtrFl = "../data/imgs_exif_data_full.json","../data/full_gid_aid_map.json","../data/full_aid_features.json"


    

In [126]:

    

# obj = PE.trainLearningObj("../data/BeautyFtrVector_GZC_Expt2.csv",
#                          "../data/GZC_exifs_beauty_full.csv",
#                          'logistic',
#                          'beauty',
#                           None,
#                           methArgs,
#                           True
#                          )

# obj, res = PE.trainTestRgrs("../FinalResults/ImgShrRnkListWithTags.csv",
#                          "../data/full_gid_aid_ftr_agg.csv",
#                          'linear',
#                          'non_sparse',
#                           None,
#                           regrArgs
#                          )
obj, res = PE.trainTestRgrs("../data/BeautyFtrVector_GZC_Expt2.csv",
                         "../data/GZC_exifs_beauty_full.csv",
                         'linear',
                         'beauty',
                          None,
                          regrArgs
                         )


    

    




Number of outliers identified: 1
9405 9405

In [124]:

    

a, results = PE.trainTestClf("../data/BeautyFtrVector_GZC_Expt2.csv",
                         "../data/GZC_exifs_beauty_full.csv",
                             'logistic',
                             'beauty',
                              None,
                              methArgs
                            )


    

In [125]:

    

res


    

    
Out[125]:





{'8287': 49.216683559356369,
 '8259': 51.854799518005493,
 '806': 40.976623978089968,
 '2029': 49.619708806657179,
 '2089': 46.91469371258566,
 '8764': 53.074001804819886,
 '4206': 44.781509699096588,
 '3950': 46.439990824972845,
 '7793': 48.08013391376808,
 '1205': 55.807458747326031,
 '1575': 56.469391803409394,
 '2362': 53.549256487678107,
 '3621': 54.431403786405959,
 '5142': 64.005551625247108,
 '8047': 46.918979677014065,
 '1311': 56.47662729294936,
 '7425': 45.055601212925907,
 '6991': 50.176266220033874,
 '265': 53.703759804079851,
 '7655': 41.568261282006048,
 '1865': 49.081933678705269,
 '2407': 62.310867355902523,
 '7022': 54.519277289121412,
 '1239': 49.293419965751269,
 '8955': 51.701047107778535,
 '2380': 55.031108928605136,
 '3563': 43.178152964248142,
 '8027': 47.497593612958433,
 '2850': 51.847676735639205,
 '4329': 40.79933072318066,
 '276': 58.998472733061071,
 '8455': 46.22894308655011,
 '1075': 51.401982340634994,
 '4811': 42.097632620371613,
 '8783': 47.513556972753207,
 '3450': 60.045920481732907,
 '1941': 51.99874905592808,
 '987': 26.501078139184106,
 '732': 45.43411124969878,
 '6672': 34.109162979233702,
 '8688': 48.55274663928742,
 '9234': 52.747798050868298,
 '1765': 50.026341509826047,
 '7407': 52.790771995265999,
 '4509': 40.843438889356527,
 '6898': 59.965155708843604,
 '1900': 51.479709649809976,
 '6576': 50.756857072902193,
 '3990': 62.316809655570488,
 '104': 48.018014364603232,
 '5598': 54.629779228309367,
 '1003': 46.082826733520903,
 '4637': 47.546254776110715,
 '113': 58.704071909474536,
 '7269': 42.902331247326281,
 '6218': 38.919911980297016,
 '5222': 51.567459255156351,
 '5190': 53.487921612102753,
 '1666': 56.897039030501858,
 '8705': 49.360685893765037,
 '8333': 53.088128926331592,
 '4798': 40.806353447294462,
 '2519': 71.137023238937957,
 '282': 52.95088105001858,
 '5466': 50.292724845006546,
 '2456': 52.794642586256288,
 '5722': 52.527840693393983,
 '2995': 54.748858531909264,
 '9361': 56.548408669550028,
 '79': 47.899982877724838,
 '7257': 50.508477768410245,
 '6572': 42.218024396704877,
 '1754': 53.796699948922381,
 '3444': 54.650575570917596,
 '8153': 42.518890402856734,
 '407': 49.308739431857845,
 '1248': 52.466220119594482,
 '1612': 48.649841553767445,
 '1053': 45.333368958834626,
 '1280': 62.26422751723036,
 '7598': 39.365802805844268,
 '130': 61.75058431847846,
 '6084': 59.994626402116744,
 '8496': 50.262311207108439,
 '3127': 50.759608493454834,
 '1516': 57.142013677641344,
 '855': 36.734869592047197,
 '5390': 53.329593478110397,
 '7357': 48.00327231075245,
 '5251': 46.206800044694702,
 '2531': 66.729519739513421,
 '2712': 51.817760340950429,
 '2328': 65.122114271577061,
 '5355': 48.621965456772514,
 '7146': 49.812783210057191,
 '2360': 71.786996543033766,
 '5269': 53.02901264424743,
 '3112': 50.07945487934019,
 '3126': 52.67602979331906,
 '8382': 54.819421174121807,
 '2294': 51.063277391241655,
 '8277': 52.607648919785078,
 '5290': 46.732192785256153,
 '3233': 54.432098592801346,
 '4702': 40.913426243267466,
 '7470': 62.667848372776689,
 '4649': 48.482534389283245,
 '9304': 52.597021837025459,
 '786': 42.497079958127799,
 '1102': 53.745381254194101,
 '527': 39.348063998527557,
 '1812': 48.207277524532934,
 '6564': 45.272240367927637,
 '8156': 43.681406568695252,
 '5913': 37.296283122768322,
 '1396': 51.501640636947485,
 '5172': 53.766088374400624,
 '8182': 48.320173793760233,
 '2849': 52.838517249337023,
 '1839': 47.258180852176466,
 '2957': 44.880409886993036,
 '2633': 57.217630993520167,
 '8173': 48.987672128504713,
 '1539': 54.678977436428639,
 '4949': 54.294270563146419,
 '764': 41.064316418501981,
 '5759': 55.548456764436239,
 '3601': 54.017987407457674,
 '5535': 46.459412565365412,
 '6387': 49.012467520092798,
 '4776': 49.647779728507516,
 '2787': 51.617863656776599,
 '83': 51.002313400839292,
 '7186': 46.738289205656429,
 '8529': 58.565869201881242,
 '5475': 48.36475961239546,
 '59': 55.490958747685582,
 '3419': 62.370664192061753,
 '4532': 46.669264262276855,
 '5866': 43.638582745507193,
 '6022': 58.142514205186899,
 '7016': 61.52630828964579,
 '8020': 47.035053043291306,
 '3981': 51.851972367692483,
 '8115': 45.518785332002466,
 '797': 42.841333238183928,
 '3316': 53.756559116533474,
 '1066': 50.082465996299675,
 '5279': 56.479507923135309,
 '5047': 55.504833558455275,
 '9018': 51.245887154123665,
 '1813': 47.88212991407449,
 '5521': 49.802522338261291,
 '5069': 51.530696269945238,
 '1932': 46.828272511207459,
 '8835': 56.225006247586123,
 '8972': 53.710951486979461,
 '7573': 43.473340576438375,
 '486': 39.926664803433916,
 '353': 58.457264212072964,
 '8218': 52.504954632951268,
 '2816': 51.964248980537597,
 '8085': 44.376467297044222,
 '5628': 51.726942009325676,
 '7950': 47.374379149034183,
 '1573': 55.351190688895016,
 '5070': 48.540638274075128,
 '6031': 41.118937730945383,
 '4498': 53.815182752984825,
 '3309': 53.938100380293946,
 '5807': 59.687604010112878,
 '2521': 69.079147688362468,
 '3169': 49.485002576285638,
 '4900': 50.95674991423509,
 '5024': 51.003190113407399,
 '7885': 47.357202597201841,
 '8084': 42.839728708729901,
 '1662': 55.465340237025018,
 '7333': 51.119702759526028,
 '2175': 61.967757895531449,
 '308': 58.102214078720344,
 '722': 42.306572608388713,
 '3397': 50.062954091818781,
 '6871': 55.343697096218385,
 '6095': 50.030874775184671,
 '3429': 61.251954267915288,
 '9110': 49.267911270585444,
 '9406': 58.212561007583588,
 '4952': 47.976940117842815,
 '2750': 54.017017182205358,
 '4069': 31.349507664509972,
 '7807': 49.729141703355801,
 '5942': 46.45582913776871,
 '637': 41.901510113492584,
 '8423': 50.576891426583416,
 '136': 58.989416157734198,
 '7654': 39.337084183408379,
 '2047': 45.843527729700149,
 '2907': 42.666850697650489,
 '4314': 39.760046039863219,
 '9189': 51.603478092373422,
 '1966': 48.684227300164999,
 '5577': 56.495788526284144,
 '1963': 49.004492445151833,
 '6698': 44.180604615055948,
 '7160': 55.477810015958475,
 '6006': 46.992796099975266,
 '5894': 42.782293884331402,
 '6018': 46.145075629543889,
 '1106': 40.905092520886292,
 '9384': 45.398246340185992,
 '1375': 53.371716429771624,
 '8632': 51.352278343105411,
 '5122': 55.709962824224988,
 '1952': 50.836728746018665,
 '3431': 51.763250165427863,
 '7579': 42.779341395248345,
 '1875': 47.716097233305732,
 '6044': 46.643410126168085,
 '7537': 46.956042812752209,
 '7669': 40.825261080885312,
 '6684': 54.156820660145577,
 '880': 47.811224509008582,
 '9038': 54.405104763820397,
 '6613': 48.985026437235497,
 '4110': 56.479595196069759,
 '7939': 48.118595869992596,
 '4242': 46.198146704985341,
 '6357': 48.112482029002081,
 '8821': 49.05783824432325,
 '6047': 44.253421732888903,
 '423': 40.231525170868579,
 '1556': 50.706966361962436,
 '3409': 54.213315950973872,
 '6858': 58.374221743703004,
 '6153': 50.622205265243835,
 '6860': 55.32273400106169,
 '6872': 57.277742022189912,
 '6678': 44.933308055976106,
 '6135': 49.64417324525806,
 '2611': 48.338751818002208,
 '8693': 48.301008672506853,
 '3203': 49.471819422329062,
 '1863': 47.352946659604854,
 '611': 38.56896162659482,
 '5720': 60.768495997534643,
 '2747': 47.710041027776597,
 '4812': 43.430557967696295,
 '2680': 49.001187073083941,
 '8755': 49.952470226113149,
 '9117': 49.95330000190318,
 '1529': 52.891076425442208,
 '8485': 53.415706118471405,
 '5828': 59.298285595012317,
 '1293': 55.20984474549924,
 '6558': 48.63060095102162,
 '7104': 49.62330339371303,
 '7985': 40.296420934834742,
 '5923': 34.282236636011021,
 '8273': 50.604746559613865,
 '7439': 50.08828852765744,
 '6922': 52.603753075565791,
 '8424': 50.329586990920049,
 '1500': 62.440222192412733,
 '2987': 57.023284761928494,
 '7388': 45.86308587665426,
 '7724': 50.953996261271541,
 '8752': 43.384123640356229,
 '3280': 53.943062951446791,
 '7769': 55.23491374198187,
 '1324': 54.847531105785869,
 '9064': 47.521689232029971,
 '1094': 54.003803349175548,
 '5498': 50.005409835232228,
 '9341': 48.032817530537784,
 '761': 43.544874200807236,
 '5236': 52.658707873081468,
 '319': 39.624746359415433,
 '3532': 39.510520902915651,
 '3264': 49.9649955930754,
 '7351': 61.18500310096762,
 '6809': 46.073263949860753,
 '9377': 48.375390976562102,
 '7843': 50.615343907885162,
 '6785': 67.185317460332129,
 '2256': 48.101646848833312,
 '1634': 51.170478770424339,
 '4569': 47.218394868798576,
 '9197': 45.288439668459276,
 '7069': 55.433102454053589,
 '6704': 61.197758094392483,
 '9366': 41.398859467078331,
 '2716': 46.213494337699373,
 '6911': 56.511775220316416,
 '4597': 51.715998910199247,
 '2074': 40.68744997084675,
 '7288': 46.021420544544377,
 '1457': 50.73375825368349,
 '5725': 62.708062745152247,
 '7404': 40.732915011378289,
 '7010': 52.650194045267241,
 '2636': 47.166535472488356,
 '2377': 53.660269909480661,
 '9364': 50.526086164329257,
 '2431': 56.458335760400274,
 '6406': 47.155799237487543,
 '3676': 59.295031699570387,
 '5754': 60.33346217307782,
 '4738': 52.783872524394212,
 '2015': 47.480772854316079,
 '3731': 52.871265420781405,
 '6526': 42.134157356539966,
 '3068': 66.28896319559091,
 '6025': 50.350737223181497,
 '8434': 51.321610716814341,
 '1772': 48.025907668964791,
 '365': 50.397732833159566,
 '5950': 39.377576809203077,
 '4679': 46.462717975174449,
 '7789': 50.321321158621565,
 '4196': 44.408116862033665,
 '6015': 43.432856600112281,
 '3598': 53.025282999385084,
 '525': 54.637988628677199,
 '6160': 47.487279638056108,
 '9247': 49.833314454739266,
 '3398': 52.363565568915774,
 '4694': 63.880614505492019,
 '369': 55.503908941976782,
 '1740': 49.022181884885867,
 '7108': 51.747205663226239,
 '7784': 45.774342597609248,
 '4641': 46.799422522826021,
 '998': 41.452372341927735,
 '7788': 46.454352342050548,
 '9255': 50.741787587577079,
 '1227': 41.95440913848978,
 '6391': 46.865672053282736,
 '3960': 47.652043353230709,
 '7475': 55.872183959745598,
 '5961': 39.839805688328042,
 '6434': 64.505212242013243,
 '1262': 50.16506136743174,
 '1661': 53.510067805453602,
 '2120': 66.759463042755115,
 '7542': 47.69385938621695,
 '8512': 52.079866751777239,
 '7117': 45.517637763468016,
 '943': 37.1886559546911,
 '2033': 45.587013742590479,
 '4028': 45.563430841868019,
 '3109': 53.709980314927378,
 '7481': 61.998007428706401,
 '2862': 52.410411367732188,
 '8457': 52.279597609545647,
 '2572': 52.569607121951442,
 '1498': 59.680317142835463,
 '4423': 56.052938427400186,
 '3643': 50.283945996584222,
 '2983': 59.074148758326189,
 '3665': 52.402254446428152,
 '9017': 46.090889866927419,
 '3364': 51.943179473478786,
 '8141': 46.662117171441402,
 '6887': 60.667358508610754,
 '6940': 58.054383989421289,
 '3462': 53.236338032044536,
 '4359': 28.092506517238238,
 '8918': 50.67592677541716,
 '8087': 47.678996214384895,
 '1246': 43.845475911848453,
 '9305': 53.203934193647179,
 '1050': 53.476772012723337,
 '2711': 47.547914044261425,
 '4622': 47.96469593455668,
 '4610': 53.304020825278798,
 '426': 49.495263218087288,
 '4918': 46.051673988293658,
 '4822': 46.39113368688124,
 '9154': 49.22396161149959,
 '8039': 44.441816669728773,
 '4659': 48.933199891270576,
 '3982': 56.011694941329665,
 '810': 40.717473004276194,
 '8036': 42.723374649391403,
 '6841': 60.780520948396877,
 '9357': 60.919971938902265,
 '443': 61.905093042545388,
 '7823': 44.009041367094653,
 '2963': 46.817089713889928,
 '3799': 59.710502143939564,
 '4073': 29.600781143284877,
 '6224': 48.106957475477536,
 '6679': 53.967442204041866,
 '4188': 48.622451452157122,
 '2166': 59.747306741765058,
 '5826': 56.521928895730284,
 '4652': 47.987917685911007,
 '3919': 54.123101204789847,
 '5712': 51.659269427493236,
 '2671': 47.102174109800835,
 '6244': 46.483314591794844,
 '1533': 59.856254708039586,
 '3701': 52.405504683087393,
 '135': 50.554256072425169,
 '3325': 71.554084433918348,
 '5994': 49.85744658294513,
 '1541': 54.479293906789991,
 '4492': 44.792285069162737,
 '4878': 40.800568259811733,
 '6626': 44.915678701073155,
 '5447': 50.014191895202671,
 '7020': 49.466991399273333,
 '3622': 55.716303749929047,
 '7167': 48.097352188923644,
 '1499': 62.324990884483256,
 '6053': 43.324463001179701,
 '6331': 48.668978701650488,
 '3168': 64.307038976070544,
 '1637': 55.156816021968709,
 '4644': 47.56030952390126,
 '2870': 55.774655675182572,
 '2688': 53.088844373817778,
 '424': 41.557515729859148,
 '9294': 41.493806735878529,
 '888': 58.014793668983685,
 '1345': 69.990676341222695,
 '8628': 51.716304645797024,
 '1618': 55.342722383012017,
 '7681': 43.200499585867561,
 '391': 49.90992505867402,
 '4592': 59.823986061066776,
 '8850': 45.713375765495684,
 '4273': 48.801855703534429,
 '6995': 59.691668097033038,
 '2326': 65.524049919485307,
 '971': 46.968705177321169,
 '3918': 47.672648128895602,
 '7055': 54.476597904577694,
 '4583': 49.365065681585975,
 '222': 50.425192637202549,
 '5036': 51.275838948130968,
 '2573': 57.939955125441003,
 '6390': 53.183220321003013,
 '1642': 57.31958067024614,
 '6068': 54.104113046541315,
 '8058': 46.252079860189497,
 '3966': 51.792954364614502,
 '5912': 44.392143510765095,
 '861': 51.743577257702917,
 '6946': 57.460048735035627,
 '6454': 47.50515573329389,
 '9143': 49.372142033154596,
 '9338': 49.723964201578916,
 '259': 57.236895396325913,
 '4371': 50.229698068676853,
 '7545': 45.940063385358428,
 '4668': 49.283219685397071,
 '7859': 51.723273559194801,
 '5670': 44.787884262387607,
 '9257': 47.377284184894194,
 '4479': 52.632045012855308,
 '6671': 38.045547581329849,
 '2659': 47.305013273639567,
 '2872': 51.327113485313632,
 '7638': 51.661886077692785,
 '2455': 58.296042204567911,
 '6042': 52.009110973682589,
 '1821': 49.225955713822856,
 '1218': 47.353409713646208,
 '6003': 49.082851210852176,
 '3786': 57.156344520586245,
 '2756': 53.521481706332594,
 '2234': 55.074161279947674,
 '8199': 47.675031346480694,
 '9046': 53.107012845877911,
 '3806': 45.843818477075246,
 '5605': 51.605106113710669,
 '3315': 64.606309969896344,
 '8999': 46.154335807405062,
 '347': 53.693396814445521,
 '342': 45.966380140206482,
 '8795': 48.507024012432517,
 '7441': 51.08309202174425,
 '4281': 43.079280816046705,
 '4911': 45.410646846607712,
 '3835': 46.037693322056036,
 '9196': 50.122415603058613,
 '2536': 57.550282458845984,
 '8319': 51.977193393584734,
 '1299': 63.09454041689753,
 '5504': 47.900347102803238,
 '7258': 49.367369871398097,
 '5367': 42.059534664088957,
 '1112': 61.565382847572764,
 '5363': 51.071644335011086,
 '4682': 45.299335094271058,
 '6818': 61.663612624263401,
 '9218': 49.965624568245779,
 '316': 53.879161176725546,
 '9158': 49.021937592712852,
 '5369': 42.333654511025415,
 '5066': 46.066730977358404,
 '3785': 66.801680172345584,
 '6354': 47.550676637394439,
 '4539': 44.549069012391719,
 '6278': 47.808138023278829,
 '9278': 51.523274253725155,
 '3517': 38.941888682187717,
 '7811': 46.374285420546094,
 '2150': 62.451345416691012,
 '1447': 57.632727264862453,
 '395': 47.2119810753588,
 '1625': 54.15755463837742,
 '2016': 46.902292166323605,
 '3522': 40.51900335823867,
 '4801': 39.170670477470523,
 '6571': 43.454946690284217,
 '2000': 47.652799532312706,
 '8421': 51.118795267610537,
 '1157': 58.039458738156725,
 '2010': 48.624479219743307,
 '5': 39.782141749315834,
 '3716': 52.571957043410727,
 '7882': 48.947319307176727,
 '1770': 51.281479174747879,
 '3423': 46.93183572733809,
 '6560': 42.60138919786116,
 '4749': 42.404981796974212,
 '7150': 49.989623670762754,
 '3609': 56.993339698660137,
 '2897': 44.209446655250474,
 '6430': 62.998606749365976,
 '8562': 50.231316816985256,
 '6253': 47.620878302237443,
 '6889': 55.476700120364335,
 '3078': 55.516903086836699,
 '7920': 44.646110370596439,
 '679': 51.971137462072505,
 '1332': 58.271461337721576,
 '2287': 42.97508918558492,
 '9372': 54.832061454781417,
 '124': 52.915508008031452,
 '8787': 43.989008959417248,
 '3742': 54.383355273084831,
 '9264': 51.331888292379546,
 '6402': 45.464119897428226,
 '2701': 50.773176310636146,
 '941': 50.813702953779277,
 '4391': 47.978547773101177,
 '1382': 60.229808403721634,
 '1153': 49.621504498071815,
 '875': 59.100015589810994,
 '1696': 59.452747357483645,
 '8515': 53.096800883923876,
 '6452': 38.189763862618818,
 '8387': 54.34686676460111,
 '4456': 20.580035670845014,
 '2317': 57.882664402501035,
 '9314': 52.492442831081924,
 '6397': 52.015669356889681,
 '6568': 42.884633397688361,
 '793': 40.384003185718811,
 '8804': 46.163889791059546,
 '7722': 49.170430163041154,
 '2721': 49.073359600442615,
 '7431': 54.091773207863113,
 '6071': 48.043792527306401,
 '4238': 42.123205127469667,
 '9327': 50.487832350877746,
 '476': 45.858000158587288,
 '8063': 43.662873789163385,
 '7670': 44.298544078520223,
 '2087': 39.555198528563828,
 '250': 59.531490573668378,
 '6004': 48.812290896252776,
 '7244': 44.592780474289846,
 '2767': 48.136618368082495,
 '3807': 48.283516565932715,
 '5910': 28.674596663643989,
 '6333': 47.990663083946416,
 '8638': 54.373163215464572,
 '2692': 52.451626521028118,
 '4326': 43.861603205663982,
 '6415': 42.722935468485574,
 '195': 53.635139693532267,
 '2068': 41.756528651636344,
 '858': 49.981723184318035,
 '9374': 49.744649484431726,
 '1477': 56.451655931849132,
 '5210': 51.102873744309605,
 '7790': 45.680291924794624,
 '2535': 61.835053023946266,
 '8573': 50.387078190938482,
 '1329': 56.994901181566206,
 '1316': 61.811084034081674,
 '6208': 40.967440545577176,
 '7690': 47.493111564714468,
 '1122': 52.588045773214773,
 '4173': 44.147149658666493,
 '4611': 54.385196624009083,
 '280': 56.224413559256156,
 '6640': 34.779587824057124,
 '3106': 52.071816598502664,
 '1698': 59.542199030591291,
 '6726': 41.717053056590842,
 '49': 51.833018074337076,
 '2273': 32.032320968572762,
 '4120': 44.936699245356159,
 '5584': 54.115279182085317,
 '1646': 57.839080176295369,
 '7796': 47.759493519329027,
 '8560': 46.249532303433476,
 '8715': 47.280331057410365,
 '5676': 52.647394532850846,
 '1549': 53.039431086229257,
 '4051': 48.732965319530919,
 '8789': 48.573382917478973,
 '1937': 43.021838345161861,
 '6384': 52.053663289966643,
 '1123': 45.82376667535253,
 '9187': 53.214427149591032,
 '3569': 49.805589384611757,
 '3334': 55.821981638901896,
 '8687': 49.021914122523967,
 '4096': 52.846995047004967,
 '5383': 46.994700635450513,
 '1704': 56.866486124400055,
 '7229': 49.664085409695851,
 '803': 40.471525495514079,
 '6048': 50.802476762745044,
 '6989': 50.323172960296105,
 '3353': 49.50391882404174,
 '7666': 46.057583769355197,
 '1899': 51.467689315877863,
 '2530': 67.689920348055736,
 '8240': 53.115306568794097,
 '1380': 54.175887602095429,
 '2406': 59.001310008262905,
 '2782': 48.142653788656418,
 '8393': 55.445642073696298,
 '7103': 49.318930107093749,
 '196': 52.348932279434031,
 '6119': 54.22956668265158,
 '3936': 46.009648327531075,
 '1543': 56.59954468255917,
 '2018': 45.28016182040033,
 '4588': 54.449266957427824,
 '1425': 57.980054377745986,
 '4276': 45.529618773836241,
 '2892': 44.864869091477857,
 '1934': 49.209306010353757,
 '3533': 39.135540324189854,
 '4149': 45.661049661116969,
 '431': 47.135181959629904,
 '2776': 52.783042420946572,
 '9171': 47.412856839259753,
 '6473': 59.107415250064207,
 '1349': 59.565380181564805,
 '5700': 38.137534641944242,
 '116': 63.254864533250085,
 '5904': 41.094803459297687,
 '4518': 50.359730616939302,
 '6027': 48.518455956879535,
 '7264': 49.611672328445309,
 '1038': 37.590629402229332,
 '6467': 42.784169468530052,
 '3914': 49.049247391280431,
 '3854': 45.210317295488416,
 '4026': 46.957585477411271,
 '8647': 49.07047705629212,
 '2838': 59.792933707399087,
 '436': 54.193916342421247,
 '1408': 58.40661260426517,
 '5763': 48.344272230060717,
 '2395': 68.522214207104895,
 '6239': 53.022898487296366,
 '1538': 55.105079864270834,
 '9029': 46.96308992438761,
 '8911': 48.331808049925201,
 '4650': 47.525184847343183,
 '4481': 44.803581394233142,
 '2436': 62.980920469009938,
 '9156': 45.67982375992726,
 '4187': 47.844699759682662,
 '6627': 47.481160323010641,
 '1737': 51.249986597420673,
 '5179': 49.405594033779821,
 '7661': 44.918846668778436,
 '3261': 49.899421050267051,
 '6848': 53.273186804386334,
 '7368': 53.405460465423246,
 '9140': 47.668897480199718,
 '1669': 56.543922071019153,
 '7768': 47.179357723949273,
 '6919': 56.095888262684014,
 '8315': 45.608725870290002,
 '4204': 50.390811575579683,
 '638': 39.804155753824432,
 '951': 49.770966447112727,
 '9303': 54.437769954884153,
 '5398': 59.240969153842926,
 '3130': 53.318063154792476,
 '4741': 35.96015808880329,
 '3454': 59.049658184707113,
 '1953': 51.467605013742343,
 '1448': 64.981759622911028,
 '5993': 47.860884919416122,
 '8188': 47.70307408790471,
 '2488': 59.57496507664834,
 '339': 40.678127308222528,
 '6805': 55.200089952165207,
 '5243': 54.040386890135849,
 '1650': 56.855288250409373,
 '8264': 45.66666079916245,
 '6542': 38.802586961893162,
 '7034': 46.963020248510873,
 '8091': 48.402377973811895,
 '5748': 46.185053191472321,
 '1166': 54.401727910424498,
 '7394': 44.179488226820517,
 '8016': 47.036702236303995,
 '6595': 55.910848877413372,
 '1317': 54.922232903096251,
 '1555': 49.778103842223075,
 '416': 46.055905998618208,
 '1357': 53.659724426111552,
 '9315': 51.157273401583332,
 '1570': 56.693492178799787,
 '2538': 51.036487541100684,
 '4140': 22.38453527528398,
 '9225': 48.880334402099443,
 '8621': 56.350377319013688,
 '8545': 54.580940241209611,
 '8833': 56.488037746772008,
 '4029': 45.786655666724066,
 '3069': 49.012224303372314,
 '6792': 67.824914220265526,
 '8121': 45.283011261991128,
 '6358': 48.564301732720402,
 '7833': 47.999093439765055,
 '1233': 43.649354660333579,
 '6658': 39.933738339397202,
 '2749': 49.144818005382938,
 '5470': 49.374729738937617,
 '5105': 57.430368269745671,
 '8858': 47.948686134225802,
 '9077': 46.311453038553019,
 '6747': 47.410131119333577,
 '5332': 47.831498627870097,
 '1771': 47.582785490520209,
 '5391': 52.664433086735869,
 '8316': 50.161735754356883,
 '230': 58.364151440158146,
 '7526': 53.804220801287983,
 '7411': 56.72626735673812,
 '3120': 53.839189090807302,
 '5216': 53.046479871684355,
 '9019': 45.613752373062248,
 '7556': 52.92346320547901,
 '8482': 49.117175495426316,
 '5744': 60.889554868032874,
 '2095': 66.329511587551465,
 '2442': 56.510877132755169,
 '3718': 57.787135040135311,
 '688': 39.278628175937975,
 '3244': 45.187254623018113,
 '1940': 52.058830368254796,
 '7364': 48.312828645135959,
 '5832': 56.188489348931036,
 '5231': 43.907993103354606,
 '7707': 29.711930095846476,
 '4920': 47.257897873947535,
 '6984': 57.069971245073425,
 '5063': 50.245497394307378,
 '6556': 42.273393831707622,
 '3346': 53.487158411635988,
 '780': 35.02989903535245,
 '5618': 48.828809388744368,
 '7462': 50.708753550775285,
 '6480': 57.439857680588702,
 '2174': 59.868527273242321,
 '4780': 56.376736121698777,
 '6181': 50.925176580086287,
 '1336': 63.143997856701283,
 '6023': 55.113790070767749,
 '7589': 49.258740079022338,
 '7242': 49.255681630997039,
 '5875': 45.394717625999867,
 '8062': 41.57760470111198,
 '5745': 62.838254217393484,
 '3192': 49.035633205222794,
 '6369': 43.091635088825313,
 '2410': 49.356957677084353,
 '7284': 42.808319116685297,
 '650': 44.057510476491665,
 '2358': 62.412348788823792,
 '1783': 49.216611326892945,
 '8888': 52.002307496538783,
 '2277': 44.715558707402295,
 '8753': 45.535484933787771,
 '3380': 53.783258007431918,
 '3171': 52.727351840518715,
 '3565': 46.17412578678465,
 '1800': 47.491380215961712,
 '7274': 44.860769018607044,
 '3822': 52.214913411809647,
 '7097': 47.412239192920921,
 '884': 56.598232006461174,
 '4977': 46.054826078874605,
 '8484': 52.373898229068189,
 '2709': 51.269432190090455,
 '1267': 54.113066280834964,
 '6194': 50.760041849728964,
 '3443': 50.914882043323615,
 '1433': 65.401291432986142,
 '8444': 54.549438403878192,
 '9378': 47.405273979960015,
 '9116': 49.871207337907791,
 '4166': 54.098779450802716,
 '8561': 48.016823351451208,
 '3857': 41.886471261021839,
 '1260': 54.623472369822792,
 '2934': 53.269152624453611,
 '7819': 49.093004401942103,
 '9371': 55.298835371624676,
 '2931': 51.947194230012116,
 '3362': 49.678196563415405,
 '6699': 45.49941156768255,
 '357': 49.859934998041922,
 '2726': 45.494770264119623,
 '6822': 55.033453564107788,
 '325': 50.429181640730974,
 '2657': 47.386663918011401,
 '2072': 41.822496266908146,
 '3840': 48.820340063380478,
 '7428': 48.026888636467334,
 '5783': 54.750241395936136,
 '7197': 42.973263421753941,
 '4339': 48.376853358253356,
 '1720': 53.78713706831924,
 '1960': 49.918744527846755,
 '3938': 46.959434343100277,
 '8773': 49.417271561851621,
 '604': 49.83541674201274,
 '7287': 40.581187576465979,
 '3133': 52.390798647038437,
 '4360': 16.94767433345525,
 '7561': 45.265164613549778,
 '1711': 64.637114997236907,
 '7211': 42.131820975864287,
 '842': 48.222340928856816,
 '1816': 43.046134922766711,
 '4488': 30.955548878987933,
 '6183': 51.113005946206826,
 '997': 42.142505201969151,
 '7908': 44.65753199553464,
 '7979': 48.691147949807444,
 '7940': 42.682433776216548,
 '4446': 31.430056423101309,
 '583': 50.540302261770869,
 '8681': 46.824610329375346,
 '1921': 45.280056897676289,
 '5821': 54.766958586123877,
 '466': 49.901319944072355,
 '4036': 53.12036252768975,
 '5857': 59.057480901725029,
 '8652': 48.165002910141126,
 '6203': 48.822881927890883,
 '5851': 56.876440061627392,
 '2682': 51.866561954048386,
 '533': 54.755648503758813,
 '8127': 47.135222933874168,
 '170': 50.350121345682318,
 '4973': 46.837393119268519,
 '6416': 42.609412720126677,
 '3555': 47.609526061831552,
 '5533': 50.093835248477134,
 '7233': 46.433589418008303,
 '413': 50.046058525181067,
 '2485': 57.314791054113158,
 '3559': 43.800997178954688,
 '5375': 44.628063885125123,
 '1836': 52.055190954174549,
 '9101': 48.952458751750299,
 '1393': 53.115388357372467,
 '6600': 44.67890587740699,
 '8049': 48.948566541344753,
 '2392': 51.82027239339736,
 '8987': 53.416838068384784,
 '6979': 53.024418555791947,
 '4663': 48.05448473832903,
 '7052': 49.885886925836111,
 '1790': 49.657426312547784,
 '5316': 50.263644650435218,
 '5925': 39.435329490609575,
 '6486': 41.22161595577807,
 '5921': 36.813801499935003,
 '3118': 50.16303097755987,
 '2788': 49.381232278763001,
 '5045': 50.10289045487059,
 '848': 45.290391792059964,
 '1568': 58.265919036686014,
 '5530': 46.569665823696354,
 '3197': 51.152781304563902,
 '795': 40.809914305185018,
 '908': 47.801408577662578,
 '5920': 25.987227287127673,
 '7934': 48.654227471041928,
 '5557': 37.615032138850864,
 '1522': 60.608980824662112,
 '8381': 54.660679240084647,
 '9105': 49.64286899182693,
 '6337': 50.528775556577678,
 '2479': 63.427904080236679,
 '5485': 50.318965424265627,
 '8763': 49.135802755336456,
 '3891': 49.367845939188385,
 '3236': 52.618142138665377,
 '2561': 51.252504160646211,
 '3027': 49.270484312231858,
 '5461': 49.18845499721256,
 '387': 49.487781229160959,
 '8691': 48.348715365121535,
 '2017': 47.041433318263365,
 '8710': 47.823930641199965,
 '9396': 47.556051465345874,
 '6235': 45.862139746897796,
 '8641': 53.793175284851969,
 '6429': 63.871088360803263,
 '6255': 46.239851485513327,
 '7763': 54.041014063625759,
 '5283': 48.818055163993989,
 '5989': 39.305515779678814,
 '2528': 69.565834194819161,
 '2151': 61.977556447707599,
 '4905': 51.814420904861109,
 '6863': 50.288247491934811,
 '4796': 38.960373539007598,
 '2582': 55.892097769031864,
 '1856': 50.372400374924702,
 '6113': 52.836884117099437,
 '2758': 47.847050010822407,
 '3072': 63.76555214550207,
 '6264': 45.483339806122714,
 '1415': 67.496271037732456,
 '2332': 64.795851722039743,
 '2035': 49.31364579017928,
 '144': 71.232049274563906,
 '1274': 50.722803955959094,
 '4664': 48.754883379029437,
 '3273': 49.710397878323263,
 '1167': 48.55477360026223,
 '6140': 54.166744043988047,
 '1657': 50.108717207236495,
 '6763': 61.474295067351164,
 '6565': 38.453730835866295,
 '3370': 49.962563861269984,
 '839': 43.600594449575908,
 '1247': 51.894344348537309,
 '143': 69.167371931141389,
 '8272': 49.551965293287743,
 '190': 54.967761462534263,
 '8645': 53.5473301470427,
 '266': 54.296785929595217,
 '1082': 39.099988442481987,
 '6952': 53.907596745765389,
 '5947': 42.525151769271567,
 '8219': 49.805260228895079,
 '1512': 61.885361408032153,
 '6941': 60.992854827636222,
 '1351': 59.702041367646942,
 '5246': 44.077830441609038,
 '439': 55.992465957468369,
 '3986': 50.952692327379495,
 '3138': 57.957412719195183,
 '5769': 63.11229099928795,
 '8492': 52.318122825636387,
 '8251': 46.437364133190997,
 '6961': 49.18407001562025,
 '7512': 57.525422878530648,
 '4207': 49.353170096995427,
 '7753': 48.096673716460359,
 '8994': 50.815634891495279,
 '915': 52.933437105522081,
 '1913': 46.039058339891028,
 '8855': 44.394830538215601,
 '7047': 47.999589132557084,
 '3388': 50.145209231326312,
 '3046': 40.683355516623955,
 '1430': 60.267828309538977,
 '109': 55.813134200011248,
 '751': 47.102268253081455,
 '8462': 54.52461687252449,
 '3577': 50.629050544737744,
 '5640': 51.260746801069885,
 '5736': 53.038439148261006,
 '8337': 58.896420450455651,
 '2661': 48.81478957026836,
 '7903': 54.028775960092858,
 ...}

In [127]:

    

PE.kSharesPerContribAfterCoinFlip(res,inExifFl,inGidAidMapFl,inAidFtrFl, lambda : 70)


    

    
Out[127]:





{'all': 3252.0, 'giraffes': 195.8, 'zebras': 3063.315789473684}

In [118]:

    

len(obj.preds)


    

    
Out[118]:





6469

In [162]:

    

df, titleSuffix = PE.buildErrPlots('clf')


    

In [156]:

    

df, titleSuffix = PE.buildErrPlots('clf')
attribTypes = ['beauty']
figs = []
for attrib in attribTypes:
    attribCol = [col for col in df.columns if attrib in col]
    attribDf = df[attribCol]
    titleAttrib = "All %s %s" %(attrib,titleSuffix)
    figs.append(attribDf.iplot(kind='line',title=titleAttrib))

figCodes = [fig.embed_code for fig in figs]


    

In [151]:

    

titleSuffix


    

    
Out[151]:





'classifiers top k choices'

In [157]:

    

figCodes


    

    
Out[157]:





['<iframe id="igraph" scrolling="no" style="border:none;" seamless="seamless" src="https://plot.ly/~smenon8/497.embed" height="525px" width="100%"></iframe>']

In [167]:

    

for col in df.columns:
    if 'all' in col:
        df[str(col+'_err')] = (3620 - df[col]) / 36.20


    

In [168]:

    

df.head()


    

    
Out[168]:






  

    

      

      
all_bayesian_beauty
      
giraffes_bayesian_beauty
      
zebras_bayesian_beauty
      
all_logistic_beauty
      
giraffes_logistic_beauty
      
zebras_logistic_beauty
      
all_svm_beauty
      
giraffes_svm_beauty
      
zebras_svm_beauty
      
all_dtree_beauty
      
...
      
zebras_random_forests_beauty
      
all_ada_boost_beauty
      
giraffes_ada_boost_beauty
      
zebras_ada_boost_beauty
      
all_bayesian_beauty_err
      
all_logistic_beauty_err
      
all_svm_beauty_err
      
all_dtree_beauty_err
      
all_random_forests_beauty_err
      
all_ada_boost_beauty_err
    
    

      
num_images
      

      

      

      

      

      

      

      

      

      

      

      

      

      

      

      

      

      

      

      

      

    
  
  

    

      
2
      
649.000000
      
8.0
      
1071.0
      
649.000000
      
8.0
      
1071.0
      
649.000000
      
8.0
      
1071.0
      
649.000000
      
...
      
1071.0
      
649.000000
      
8.0
      
1071.0
      
82.071823
      
82.071823
      
82.071823
      
82.071823
      
82.071823
      
82.071823
    
    

      
3
      
1232.500000
      
12.0
      
2044.0
      
1232.500000
      
12.0
      
2044.0
      
1232.500000
      
12.0
      
2044.0
      
1232.500000
      
...
      
2044.0
      
1232.500000
      
12.0
      
2044.0
      
65.953039
      
65.953039
      
65.953039
      
65.953039
      
65.953039
      
65.953039
    
    

      
4
      
2160.000000
      
14.0
      
3666.0
      
2160.000000
      
14.0
      
3666.0
      
2160.000000
      
14.0
      
3666.0
      
2160.000000
      
...
      
3666.0
      
2160.000000
      
14.0
      
3666.0
      
40.331492
      
40.331492
      
40.331492
      
40.331492
      
40.331492
      
40.331492
    
    

      
5
      
3219.500000
      
17.0
      
5520.0
      
3219.500000
      
17.0
      
5520.0
      
3219.500000
      
17.0
      
5520.0
      
3219.500000
      
...
      
5520.0
      
3219.500000
      
17.0
      
5520.0
      
11.063536
      
11.063536
      
11.063536
      
11.063536
      
11.063536
      
11.063536
    
    

      
6
      
3107.333333
      
20.0
      
4002.0
      
3107.333333
      
20.0
      
4002.0
      
3107.333333
      
20.0
      
4002.0
      
3107.333333
      
...
      
4002.0
      
3107.333333
      
20.0
      
4002.0
      
14.162063
      
14.162063
      
14.162063
      
14.162063
      
14.162063
      
14.162063
    
  

5 rows × 24 columns

In [1]:

    

import pandas as pd
import json
import re
from datetime import datetime


    

In [52]:

    

with open("../data/Flickr_EXIF_full.json", "r") as exiffl:
    d = json.load(exiffl)
df_exif = pd.DataFrame.from_dict(d).transpose()


    

In [53]:

    

with open("../data/beautyFeatures_FlickrExtracts.json", "r") as btyFl:
    dBty = json.load(btyFl)

# dBty_mod = {}
# for img in dBty.keys():
#     mod_img = re.findall(r'([0-9]*)_.*',img)[0]
#     dBty_mod[mod_img] = dBty[img]

df_bty = pd.DataFrame.from_dict(dBty).transpose()


    

In [54]:

    

df_exif.reset_index(inplace=True)
df_bty.reset_index(inplace=True)


    

In [55]:

    

df = pd.DataFrame.merge(df_exif, df_bty)
df.index = df['index']


    

In [56]:

    

df['date'] = df['date'].apply(lambda x : datetime.strptime(str(x),'%Y-%m-%d %H:%M:%S'))
df['day'] = df.date.apply(lambda x : x.day)
df['hour'] = df.date.apply(lambda x : x.hour)
df.drop(['index', 'date'],1,inplace=True)


    

In [57]:

    

df.to_csv("../data/Flickr_Scrapes_Ftrs.csv")


    

In [1]:

    

import PopulationEstimatorAPI as PE


    

In [2]:

    

train_fl = "../data/BeautyFtrVector_GZC_Expt2.csv"
test_fl = "../data/Flickr_Scrapes_Ftrs.csv"
meth = 'lasso'
attrib = 'beauty'
methArgs = {'linear' : {'fit_intercept' : True},
            'ridge' : {'fit_intercept' : True},
            'lasso' : {'fit_intercept' : True},
            'elastic_net' : {'fit_intercept' : True},
            'svr' : {'fit_intercept' : True},
            'dtree_regressor' : {'fit_intercept' : True}}
methObj,predResults = PE.trainTestRgrs(train_fl,
                                test_fl,
                                meth,
                                attrib,
                                infoGainFl="../data/infoGainsExpt2.csv",
                                methArgs = methArgs
                                )


    

    




Number of outliers identified: 0
1701 1701

In [3]:

    

max(predResults.values()), min(predResults.values())


    

    
Out[3]:





(78.140077977527866, 31.369773817297357)

In [9]:

    

imgs = list(methObj.test_x.index)


    

In [11]:

    

with open("../data/fileURLS.dat") as urlFls:
    urls = urlFls.read().split("\n")


    

In [28]:

    

flUrlDict = {}
for img in imgs:
    for url in urls:
        if str(img) in url:
            flUrlDict[str(img)] = url


    

In [29]:

    

with open("Flickr_FL_URL_map.json", "w") as fl:
    json.dump(flUrlDict, fl, indent = 4)


    

In [9]:

    

df = pd.DataFrame.from_csv("../data/Flickr_Scrapes_Ftrs.csv")


    

In [11]:

    

def rotation_to_orientation(theta):
    if theta == 0:
        return 1
    elif theta == 90:
        return 8
    else:
        return 6


    

In [12]:

    

df['orientation'] = df['rotation'].apply(rotation_to_orientation)


    

In [13]:

    

df.drop(['lat', 'long', 'day', 'rotation'], 1, inplace=True)


    

In [15]:

    

df.to_csv("/tmp/testing_fl.csv")


    

In [17]:

    

with open("../data/bing_img_data.json", "r") as bing_data:
    exif = json.load(bing_data)


    

In [25]:

    

with open("../data/beautyFeatures_Bing.json", "r") as bing_bty:
    bty = json.load(bing_bty)


    

In [28]:

    

bty = {re.findall(r'(.*).jpeg', img)[0] : bty[img] for img in bty.keys() if bty[img] != None}


    

In [36]:

    

df_bty = pd.DataFrame(bty).transpose()
df_exif = pd.DataFrame(exif).transpose()

df_bty.reset_index(inplace=True)
df_exif.reset_index(inplace=True)


    

In [39]:

    

df_bing = pd.merge(df_bty, df_exif)


    

In [51]:

    

def hour_getter(dt):
    if len(dt) != len('2001-01-01'):
        return datetime.strptime(dt,'%Y-%m-%dT%H:%M:%S').hour
    else:
        return 12


    

In [44]:

    

count


    

    
Out[44]:





29

In [52]:

    

df_bing['hour'] = df_bing['date'].apply(hour_getter)


    

In [57]:

    

df_bing.index=df_bing['index']
df_bing['orientation'] = 1

df_bing.drop(['index'],1,inplace=True)


    

In [61]:

    

df_bing.to_csv("/tmp/testing_fl_bing.csv")


    

In [59]:

    

df_bing['orientation'] = 1


    

In [15]:

    

exifDf


    

    
Out[15]:






  

    

      

      
arousal
      
contrast
      
dominance
      
hsv_itten_std_h
      
hsv_itten_std_s
      
hsv_itten_std_v
      
pleasure
      
symmetry
    
  
  

    

      
1
      
-0.105433
      
1.70422
      
0.506566
      
33183.8
      
66438.4
      
32484.3
      
0.45014
      
8.38616
    
    

      
10
      
-0.00690542
      
1.68977
      
0.471749
      
65048.1
      
67045.8
      
71687.8
      
0.410388
      
5.88898
    
    

      
100
      
0.0587454
      
3.70982
      
0.29646
      
12937.7
      
42971.5
      
64439
      
0.251809
      
13.2125
    
    

      
1000
      
0.10483
      
2.33853
      
0.434643
      
36115.3
      
29319.6
      
74245.7
      
0.367374
      
9.01152
    
    

      
1001
      
-0.136607
      
1.27727
      
0.670583
      
22162
      
40191.3
      
55682.2
      
0.595601
      
10.0584
    
    

      
1002
      
0.00611701
      
2.06173
      
0.484616
      
41980
      
55250.9
      
20080.1
      
0.420306
      
8.93633
    
    

      
1003
      
-0.0892074
      
1.85724
      
0.423546
      
13970.1
      
63587.7
      
22734.5
      
0.376468
      
8.81148
    
    

      
1004
      
-0.00307191
      
2.25071
      
0.449165
      
34241.9
      
51699.3
      
6992.05
      
0.390404
      
9.05704
    
    

      
1005
      
0.0159034
      
2.32259
      
0.372574
      
38778
      
52507
      
43100.2
      
0.322052
      
8.46487
    
    

      
1006
      
0.0238853
      
2.66009
      
0.396468
      
40227.5
      
58097.2
      
45527.2
      
0.342033
      
9.14275
    
    

      
1007
      
0.0972089
      
1.7106
      
0.659343
      
43723.4
      
30913.5
      
48908.5
      
0.563265
      
6.94748
    
    

      
1008
      
-0.0484954
      
2.37069
      
0.370191
      
19301.9
      
69451.5
      
34876.9
      
0.326199
      
13.5827
    
    

      
1009
      
0.0301359
      
2.01839
      
0.519714
      
103365
      
126539
      
64566.5
      
0.44847
      
13.4438
    
    

      
101
      
0.087114
      
3.05963
      
0.367281
      
25417.8
      
40518
      
70229.6
      
0.31058
      
13.306
    
    

      
1010
      
-0.0418707
      
1.96768
      
0.48041
      
25773.4
      
53264.2
      
50839.6
      
0.421286
      
11.0411
    
    

      
1011
      
-0.0298001
      
2.02743
      
0.45825
      
52475.5
      
57814.1
      
24656.1
      
0.400874
      
9.55014
    
    

      
1012
      
-0.10267
      
1.22848
      
0.490417
      
6928.72
      
48453.2
      
51913.3
      
0.435847
      
7.48271
    
    

      
1013
      
-0.101991
      
1.66353
      
0.510121
      
31726.3
      
51976.8
      
30434.9
      
0.452895
      
12.37
    
    

      
1014
      
0.0281162
      
2.23852
      
0.423892
      
54498
      
84437.8
      
170600
      
0.365443
      
8.63952
    
    

      
1015
      
-0.0824027
      
2.02468
      
0.39351
      
21668.7
      
64844.1
      
34235.8
      
0.349725
      
14.1669
    
    

      
1016
      
-0.0803648
      
1.9234
      
0.455232
      
32498.2
      
78735.5
      
9144.28
      
0.403135
      
10.4705
    
    

      
1017
      
0.0291234
      
1.71589
      
0.537191
      
52909.4
      
32470.5
      
41264.8
      
0.463747
      
9.78529
    
    

      
1018
      
0.136636
      
2.2447
      
0.543746
      
43463.3
      
11230.8
      
51907.2
      
0.459061
      
7.63129
    
    

      
1019
      
0.0221544
      
3.17119
      
0.310418
      
36815.8
      
118841
      
109799
      
0.267464
      
9.15517
    
    

      
102
      
0.063631
      
2.0796
      
0.493213
      
45955.6
      
47840.5
      
109469
      
0.42222
      
8.54146
    
    

      
1020
      
-0.0926312
      
1.64462
      
0.552838
      
49543.6
      
64642.5
      
57103.4
      
0.489092
      
10.287
    
    

      
1021
      
0.0603003
      
1.66671
      
0.657366
      
58240
      
63585.2
      
58620.7
      
0.565111
      
7.49182
    
    

      
1022
      
-0.0791019
      
1.46573
      
0.637612
      
40517.8
      
55800.7
      
59146.1
      
0.561413
      
9.27507
    
    

      
1023
      
0.0328336
      
2.63559
      
0.399029
      
16846.6
      
35917.1
      
35136.9
      
0.343393
      
6.9651
    
    

      
1024
      
-0.0232768
      
1.7887
      
0.513271
      
43880.2
      
62257.6
      
22821.8
      
0.448031
      
9.48204
    
    

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

      
972
      
0.0335071
      
2.60452
      
0.400976
      
37566.8
      
66517.7
      
48338.8
      
0.345019
      
8.55462
    
    

      
973
      
-0.0297146
      
2.22999
      
0.440352
      
25538.6
      
66598.7
      
7848
      
0.385321
      
11.9913
    
    

      
974
      
0.0365995
      
2.1488
      
0.479996
      
25327.9
      
44023.6
      
11284.4
      
0.413351
      
8.14082
    
    

      
975
      
0.0360703
      
5.18762
      
0.22152
      
25500.6
      
46914
      
60115.5
      
0.188911
      
9.76651
    
    

      
976
      
-0.036398
      
2.34377
      
0.404194
      
22249.3
      
65911.5
      
76065.6
      
0.354563
      
9.39296
    
    

      
977
      
-0.0643503
      
2.13917
      
0.401818
      
19483.6
      
67047
      
18791.1
      
0.355198
      
13.1596
    
    

      
978
      
0.00740882
      
1.97213
      
0.4909
      
52697.7
      
49608.9
      
48447.1
      
0.42564
      
14.1276
    
    

      
979
      
0.0115993
      
1.98573
      
0.513035
      
33367.2
      
47707.3
      
36313.1
      
0.444459
      
11.7066
    
    

      
98
      
-0.0657953
      
2.1802
      
0.382199
      
24997.3
      
56207.5
      
35936.7
      
0.338297
      
7.48948
    
    

      
980
      
-0.103927
      
2.18687
      
0.377509
      
25148.8
      
89428
      
91274.6
      
0.337906
      
7.11044
    
    

      
981
      
0.0806828
      
2.393
      
0.485996
      
49568.4
      
44741.2
      
19696.9
      
0.414306
      
10.5748
    
    

      
982
      
0.113545
      
2.37652
      
0.480274
      
50398.6
      
32646.7
      
23616.6
      
0.406163
      
15.5173
    
    

      
983
      
0.155739
      
2.4281
      
0.502711
      
46692.9
      
30726.1
      
63575.6
      
0.421577
      
9.53897
    
    

      
984
      
-0.129773
      
1.23268
      
0.658119
      
45843.1
      
54187.2
      
99763.1
      
0.584115
      
11.2731
    
    

      
985
      
-0.00586724
      
2.05155
      
0.468567
      
32131.8
      
51079.3
      
13771.6
      
0.407524
      
9.81283
    
    

      
986
      
0.0671683
      
2.1421
      
0.51498
      
51174.5
      
37452.7
      
35296.6
      
0.440784
      
9.50483
    
    

      
987
      
0.0959526
      
3.58665
      
0.333311
      
19043.7
      
31073.2
      
51141.7
      
0.280223
      
9.55091
    
    

      
988
      
0.000704361
      
2.09434
      
0.47572
      
54154.7
      
63072.7
      
22286.1
      
0.413102
      
10.8835
    
    

      
989
      
-0.0820319
      
1.94498
      
0.445087
      
16250
      
71457
      
28645.7
      
0.394484
      
10.3957
    
    

      
99
      
0.0986617
      
3.07712
      
0.362204
      
18600
      
39848.4
      
51108.6
      
0.305055
      
8.42017
    
    

      
990
      
0.050416
      
2.44862
      
0.450991
      
20914.2
      
41581.5
      
58774.9
      
0.386825
      
6.2899
    
    

      
991
      
-0.0434745
      
1.82038
      
0.482703
      
37531
      
64311.3
      
76967.9
      
0.423432
      
8.46225
    
    

      
992
      
0.0259953
      
2.07012
      
0.534644
      
16002.7
      
27413.9
      
16289
      
0.461837
      
13.1651
    
    

      
993
      
0.168238
      
2.1479
      
0.686405
      
65705.7
      
54787.9
      
54272.4
      
0.579911
      
8.37257
    
    

      
994
      
-0.0668044
      
1.87914
      
0.49273
      
42094.9
      
41892.7
      
26727.7
      
0.434394
      
6.15612
    
    

      
995
      
-0.0052082
      
1.77868
      
0.517984
      
29761.7
      
43023.7
      
17649.1
      
0.450381
      
10.337
    
    

      
996
      
0.0804238
      
2.14922
      
0.502197
      
46496.8
      
41906.2
      
49548.3
      
0.428402
      
9.28581
    
    

      
997
      
0.126691
      
2.23455
      
0.549536
      
41018.1
      
48534.4
      
82063.7
      
0.46505
      
6.91324
    
    

      
998
      
-0.104799
      
1.78719
      
0.489851
      
24496.8
      
75311.8
      
19732.9
      
0.43556
      
9.57629
    
    

      
999
      
-0.0488891
      
1.95641
      
0.508191
      
39618.5
      
58379
      
35535.7
      
0.446092
      
6.67638
    
  

5401 rows × 8 columns

In [ ]: