analysis.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-
from numpy import genfromtxt
import os
import csv
import collections
import numpy as np     ## vector and matrix operations
import scipy as sp     ## grab-bag of statistical and science tools
import matplotlib.pyplot as plt     ## matplotlib - plots
import pandas as pd     ## emulates R data frames
import statsmodels.api as sm     ## scikits.statsmodels - statistics library
import patsy    ## emulates R model formulas
import math
import sklearn as skl     ## scikits.learn - machine learning library
from sklearn import mixture as sklmix
MemberfileName="Members_Y1.csv"
global claimsAns
def svmtest():
    import numpy as np
    import pylab as pl
    from sklearn import svm
    
    
    xx, yy = np.meshgrid(np.linspace(-3, 3, 500),
                         np.linspace(-3, 3, 500))
    np.random.seed(0)
    X = np.random.randn(300, 2)
    Y = np.logical_xor(X[:, 0] > 0, X[:, 1] > 0)
    
    # fit the model
    clf = svm.NuSVC()
    clf.fit(X, Y)
    
    # plot the decision function for each datapoint on the grid
    Z = clf.decision_function(np.c_[xx.ravel(), yy.ravel()])
    Z = Z.reshape(xx.shape)
    
    pl.imshow(Z, interpolation='nearest',
              extent=(xx.min(), xx.max(), yy.min(), yy.max()), aspect='auto',
              origin='lower', cmap=pl.cm.PuOr_r)
    contours = pl.contour(xx, yy, Z, levels=[0], linewidths=2,
                          linetypes='--')
    pl.scatter(X[:, 0], X[:, 1], s=30, c=Y, cmap=pl.cm.Paired)
    pl.xticks(())
    pl.yticks(())
    pl.axis([-3, 3, -3, 3])
    pl.show()
    
def _svmprocess():
    """
    用svm 處理 numpy matrix 
    """
    np.random.shuffle(claimsAns[1:,:])
    """
    array(['MemberID', 'ProviderID', 'vendor', 'pcp', 'Year', 'specialty',
       'placesvc', 'paydelay', 'LengthOfStay', 'dsfs',
       'PrimaryConditionGroup', 'CharlsonIndex', 'AgeAtFirstClaim', 'Sex',
       'DayInHospital_Y2'], 
      dtype='|S25')

    """
    import numpy as np
    claimsAns=np.load("claims_join_member_ans_cat.npy")
    Y=[]
    X=[]
    for i in range(5000):
        Y.append(int(claimsAns[1+i,-1]))
        x=[int(claimsAns[1+i,-4]),int(claimsAns[1+i,-5])]#,int(claimsAns[1+i,-6])]  
        X.append(x)
    
    X = np.asarray(X)
    Y = np.asarray(Y)
    from sklearn import svm
    kernel='rbf'
    clf = svm.SVC(kernel=kernel, gamma=2)
    clf.fit(X, Y)
    
    
    clf.predict([[2., 2.]])
    #=============
    kernel='rbf'
    clf = svm.SVC(kernel=kernel, gamma=2)
    clf.fit(X, Y)
    
    # plot the line, the points, and the nearest vectors to the plane
    pl.figure(fignum, figsize=(4, 3))
    pl.clf()
    
    pl.scatter(clf.support_vectors_[:, 0], clf.support_vectors_[:, 1], s=80,
               facecolors='none', zorder=10)
    pl.scatter(X[:, 0], X[:, 1], c=Y, zorder=10, cmap=pl.cm.Paired)
    
    pl.axis('tight')
    x_min = 0
    x_max = 10
    y_min = 0
    y_max = 10
    
    XX, YY = np.mgrid[x_min:x_max:200j, y_min:y_max:200j]
    Z = clf.decision_function(np.c_[XX.ravel(), YY.ravel()])
    
    # Put the result into a color plot
    Z = Z.reshape(XX.shape)
    
    pl.pcolormesh(XX, YY, Z > 0, cmap=pl.cm.Paired)
    pl.contour(XX, YY, Z, colors=['k', 'k', 'k'], linestyles=['--', '-', '--'],
               levels=[-.5, 0, .5])
    
    pl.xlim(x_min, x_max)
    pl.ylim(y_min, y_max)
    
    pl.xticks(())
    pl.yticks(())

def _svmpreprocess():
    """
    array(['MemberID'0, 'ProviderID'1, 'vendor'2, 'pcp'3, 'Year'4, 'specialty'5,
       'placesvc'6, 'paydelay'7, 'LengthOfStay'8, 'dsfs'9,
       'PrimaryConditionGroup'10, 'CharlsonIndex'11, 'AgeAtFirstClaim'12, 'Sex'13], 
      dtype='|S25')

    """

    #convert Categorical to Integer
    catType=catHash()
    claimsAns=np.load("claims_join_member_ans.npy")
    for i in range(int(claimsAns.shape[1])):
        fieldName=claimsAns[0][i]
        if catType.has_key(fieldName):
            for j in range(int(claimsAns.shape[0])):
                if j==0:
                    pass
                else:
                    #sraw_input("waits")
                    #print fieldName
                    #print catType[fieldName]
                    #print i,j
                    #claimsAns[j][i]
                    #print catType[fieldName][claimsAns[j][i]]
                    claimsAns[j][i]=catType[fieldName][claimsAns[j][i]]
        return claimAns

    
def _loadcvsForFeautreAnalysis(fileName=MemberfileName):
    """
    return dict for basic feature analysis
    """
    f=open(fileName)
    data=f.read()
    f.close()
    items=data.split("\r\n")
    result={}
    indexCount={}
    for i, val in enumerate(items):
        fields=val.split(",")        
        if i==0:
            for j,f in enumerate(fields):
                indexCount[j]=f
                result[f]=[]            
        else:
            for j,f in enumerate(fields):                
                result[indexCount[j]].append(f)
    return result

def _getCategoricalfeatures(DataList,DictEnable=True):
    """
    http://scikit-learn.org/stable/modules/preprocessing.html#encoding-categorical-features
    """
    s=set(DataList)
    l=list(s)
    l.sort()
    if DictEnable:
        d=dict(enumerate(l))
        res = dict((v,k) for k,v in d.iteritems())
        return res
    else:
        return l
def MergeClaim_MemeberAndAns():
    claims=np.load("claims_join_member.npy")
    Y2Ans=loadcsv("DayInHospital_Y2.csv")
    DY2=np.empty([644707,1],dtype=str)
    claims=np.column_stack([claims,DY2])
    AnsDict={}
    for i in range(Y2Ans.shape[0]):
        AnsDict[Y2Ans[i][0]]=Y2Ans[i][1]
    claims[0,-1]='DayInHospital_Y2'
    for i in range(claims.shape[0]):
        if i==0:
            pass
        else:
            claims[i][-1]=AnsDict[claims[i][0]]
    np.save("claims_join_member_ans",claims)
    
    return claims
    



def MergeMaxrix():
    """
    Try it 
    http://stackoverflow.com/questions/877479/whats-the-simplest-way-to-extend-a-numpy-array-in-2-dimensions
    """
    member=loadcsv()
    claims=loadcsv("Claims_Y1.csv")
    insertA=np.empty([644707,2],dtype=str)
    claims=np.column_stack([claims,insertA])
    MemberDict={}
    for i in range(member.shape[0]):
        MemberDict[member[i][0]]=(member[i][1],member[i][2])
    claims[0,-2]='Sex'
    claims[0,-1]='AgeAtFirstClaim'
    for i in range(claims.shape[0]):
        if i==0:
            pass
        else:
            claims[i][-1]=MemberDict[claims[i][0]][0]
            claims[i][-2]=MemberDict[claims[i][0]][1]
    
    np.save("claims_join_member",claims)  #npy file
    #claims.tofile("claims_join_member.csv",sep=",") #csv file 沒有換行
    #np.savetxt("claims_join_member.csv",claims, delimiter=",")
    claims=np.load("claims_join_member.npy")

def convertField(x,mappingList):
    return mappingList.index(x)
    

def loadcsv(fileName=MemberfileName):    
    #persontype = np.dtype({'names':['MemberID','AgeAtFirstClaim','Sex'],'formats':['i','S10', 'S1']})
    #persontype = np.dtype({'names':['MemberID','AgeAtFirstClaim','Sex'],'formats':['i','i', 'i']})
    #persontype = [('MemberID','i'),('AgeAtFirstClaim','i'),('Sex','i')]
    #dtype=[('myint','i8'),('myfloat','f8'),('mystring','S5')]
    #my_data = genfromtxt(fileName, delimiter=',',dtype= persontype,skiprows=1)
    #d=_loadcvsForFeautreAnalysis(fileName)
    #ageList=_getCategoricalfeatures(d['AgeAtFirstClaim'],False)
    #sexList=_getCategoricalfeatures(d['Sex'],False)    
    #ageConv = lambda x: ageList.index(x)
    #sexConv = lambda x: sexList.index(x)
    #np.loadtxt(fileName, delimiter=',')
    #result = genfromtxt(fileName, delimiter=',',skiprows=1,converters = {1:ageConv,2:sexConv},dtype= persontype)
    #reader=csv.reader(open(fileName,"rb"),delimiter=',')
    #result=np.loadtxt(fileName, delimiter=',',skiprows=1,converters = {1:ageConv,2:sexConv},dtype= persontype)
    #result=np.loadtxt(fileName, delimiter=',',dtype= None)
    result=np.genfromtxt(fileName, delimiter=',',dtype= None)
    #x=list(reader)
    #result=np.array(x[1:])
    
    #print ageList
    #print sexList
    
    #for i in range(result.shape[0]):
    #    result[i][1]=ageList[result[i][1]]
    #    result[i][2]=sexList[result[i][2]]
    
    return result

def catHash():
    """
    charList
    Out[52]: {'0': 0, '1-2': 1, '3-4': 2, '5+': 3}
    dict(zip([1,2,3,4], [a,b,c,d]))
    """
    result={}
    result['LengthOfStay']={'': 0,
    '1 day': 1,
    '2 days': 2,
    '3 days': 3,
    '4 days': 4,
    '5 days': 5,
    '6 days': 6,
    '1- 2 weeks': 10,
    '2- 4 weeks': 20,
    '4- 8 weeks': 42,
    '8-12 weeks': 70,
    '12-26 weeks': 140,
    '26+ weeks': 168
    }
    result['AgeAtFirstClaim']={'0-9': 0,
    '10-19': 1,
    '20-29': 2,
    '30-39': 3,
    '40-49': 4,
    '50-59': 5,
    '60-69': 6,
    '70-79': 7,
    '80+': 8}
    result['Sex']={'F': 0, 'M': 1}
    
    #['paydelay':163,'vendor':3851, 'dsfs':12, 'placesvc':8, 'pcp':1012, 'MemberID':77290,
    #'CharlsonIndex':4, 'specialty':12, 'ProviderID':7825, 'PrimaryConditionGroup':45, 'Year':1, 'LengthOfStay':13]
    val=dict(enumerate(range(0,162)))
    result['paydelay']=dict((str(v),k) for v,k in val.iteritems())  #-1,0,1,...,161 共163筆
    result['paydelay']['']=-1
    
    result['dsfs']={'0- 1 month': 0,
    '1- 2 months': 1,
    '10-11 months': 2,
    '11-12 months': 3,
    '2- 3 months': 4,
    '3- 4 months': 5,
    '4- 5 months': 6,
    '5- 6 months': 7,
    '6- 7 months': 8,
    '7- 8 months': 9,
    '8- 9 months': 10,
    '9-10 months': 11}
    
    result['placesvc']={'Ambulance': 0,
    'Home': 1,
    'Independent Lab': 2,
    'Inpatient Hospital': 3,
    'Office': 4,
    'Other': 5,
    'Outpatient Hospital': 6,
    'Urgent Care': 7}
    
    result['CharlsonIndex']={'0': 0, '1-2': 1, '3-4': 2, '5+': 3}
    result['specialty']={'Anesthesiology': 0,
    'Diagnostic Imaging': 1,
    'Emergency': 2,
    'General Practice': 3,
    'Internal': 4,
    'Laboratory': 5,
    'Obstetrics and Gynecology': 6,
    'Other': 7,
    'Pathology': 8,
    'Pediatrics': 9,
    'Rehabilitation': 10,
    'Surgery': 11}
    result['PrimaryConditionGroup']={'AMI': 0,
    'APPCHOL': 1,
    'ARTHSPIN': 2,
    'CANCRA': 3,
    'CANCRB': 4,
    'CANCRM': 5,
    'CATAST': 6,
    'CHF': 7,
    'COPD': 8,
    'FLaELEC': 9,
    'FXDISLC': 10,
    'GIBLEED': 11,
    'GIOBSENT': 12,
    'GYNEC1': 13,
    'GYNECA': 14,
    'HEART2': 15,
    'HEART4': 16,
    'HEMTOL': 17,
    'HIPFX': 18,
    'INFEC4': 19,
    'LIVERDZ': 20,
    'METAB1': 21,
    'METAB3': 22,
    'MISCHRT': 23,
    'MISCL1': 24,
    'MISCL5': 25,
    'MSC2a3': 26,
    'NEUMENT': 27,
    'ODaBNCA': 28,
    'PERINTL': 29,
    'PERVALV': 30,
    'PNCRDZ': 31,
    'PNEUM': 32,
    'PRGNCY': 33,
    'RENAL1': 34,
    'RENAL2': 35,
    'RENAL3': 36,
    'RESPR4': 37,
    'ROAMI': 38,
    'SEIZURE': 39,
    'SEPSIS': 40,
    'SKNAUT': 41,
    'STROKE': 42,
    'TRAUMA': 43,
    'UTI': 44}
    return result   







def showHistogram(data):
    data=list(data)
    correlation = [(i, data.count(i)) for i in set(data)]
    correlation.sort(key=lambda x: x[1])    
    labels, values = zip(*correlation)
    indexes = np.arange(len(correlation))
    width = 1
    plt.bar(indexes, values, width)
    plt.xticks(indexes + width * 0.5, labels)
    plt.show()


def _countFreq(items):
    """
    print(counter)
    # Counter({1: 4, 2: 4, 3: 2, 5: 2, 4: 1})
    print(counter.values())
    # [4, 4, 2, 1, 2]
    print(counter.keys())
    # [1, 2, 3, 4, 5]
    print(counter.most_common(3))
    # [(1, 4), (2, 4), (3, 2)]
    """
    counter=collections.Counter(items)
    return counter
    
def freqDist():
    """
    計算頻率分佈
    """
    import numpy as np
    import analysis as a 
    claims=np.load("claims_join_member.npy")
    nameList=list(claims[0])
    for i in range(3,14):
        print "===========",nameList[i],"==================="
        print a._countFreq(claims[1:,i])
        print "=========================================================="

def test():
    """
    1.計算feature與住院的關聯
    2.住院與不住院的差異?
    3.今年住院跟明年有關嗎?
    """
    #filter
    #c=claims[(claims[:,12]=="30-39") | (claims[:,12]=="40-49")]
    l=list(set(claims[:12]))
    l.sort()
    for i in l:
        c=claims[(claims[:,12]==i)]
    print np.corrcoef(a,b)
    
    
def summary(filePath="Claims_Y1.csv"):
    fileName, fileExtension = os.path.splitext(filePath)
    if fileExtension==".npy":
        claimsAns=np.load(filePath)    
        d=pd.DataFrame(claimsAns)
    elif fileExtension==".csv":
        d=pd.read_csv(filePath)
    d.describe()
    return d
def RMSLE_Release1():
    dihY2=pd.read_csv("DayInHospital_Y2.csv",index_col='memberid')
    allZero_RMSLE=math.sqrt(sum(dihY2['DaysInHospital_Y2'].apply(math.log1p).pow(2))/dihY2.shape[0])
    print allZero_RMSLE
    #dihY2['ans'] = pd.Series(dihY2['DaysInHospital_Y2'], index=dihY2.index)
    dihY2['ans'] = pd.Series([0]*dihY2.shape[0], index=dihY2.index)
    #print dihY2
    diff=dihY2['DaysInHospital_Y2'].apply(math.log1p)-dihY2['ans'].apply(math.log1p)    
    #print diff
    diff=diff.pow(2)
    #print diff
    print math.sqrt(sum(diff)/dihY2.shape[0])
    #print math.sqrt(sum(diff)/dihY2.shape[0])
    return dihY2

def RMSLE_Release3():
    dihY2=pd.read_csv("DayInHospital_Y2.csv",index_col='MemberID')
    allZero_RMSLE=math.sqrt(sum(dihY2['DaysInHospital_Y2'].apply(math.log1p).pow(2))/dihY2.shape[0])
    print allZero_RMSLE
    #dihY2['ans'] = pd.Series(dihY2['DaysInHospital_Y2'], index=dihY2.index)
    dihY2['ans'] = pd.Series([0]*dihY2.shape[0], index=dihY2.index)
    #print dihY2
    diff=dihY2['DaysInHospital'].apply(math.log1p)-dihY2['ans'].apply(math.log1p)    
    diff=diff.pow(2)
    print math.sqrt(sum(diff)/dihY2.shape[0])
    #math.sqrt(sum(/dihY2.shape[0])
    return dihY2


def calimsPlot():
    """
    ipython -pylab
    """
    claims=pd.read_csv("Claims_Y1.csv",index_col='MemberID')
    for i in claims.columns:        
        t=claims[i].value_counts()
        if t.count()>30:
            t.plot(title=i)
        else:
            t.plot(title=i,kind='bar',yticks=t)
        raw_input("wait...")
        plt.clf()
            
def Featurescorr():
    """
    DaysInHospital_Y215 = 0.679839894557
    DaysInHospital_Y20  = -0.679463423154
    DaysInHospital_Y28  = 0.239186703963
    DaysInHospital_Y210 = 0.237311461682
    DaysInHospital_Y214 = 0.235583859674    
    DaysInHospital_Y212 = 0.236856544715
    DaysInHospital_Y26 = 0.206533784348
    """
    from sklearn import cross_validation
    from sklearn import svm
    from sklearn.ensemble import RandomForestClassifier
    m=pd.load("featuresSet")
    result={}
    for i in range(15,299):
        fCorr=m[m.columns[i]].corr(m["DaysInHospital_Y2"])
        #if ((fCorr>0.2) or (fCorr<-0.2)):
        print m.columns[i],"=",fCorr
        result[m.columns[i]]=fCorr
        
    return result    

def Test():
    from sklearn import cross_validation
    from sklearn import svm
    from sklearn.ensemble import RandomForestClassifier
    m=pd.load("featuresSet")
    result={}
    for i in range(15,299):
        X=m[[m.columns[i]]]
        print m.columns[i]
        Y=m["DaysInHospital_Y2"]
        X_train, X_test, y_train, y_test=cross_validation.train_test_split(X,Y,test_size=0.95,random_state=0)
        clf = RandomForestClassifier(n_estimators=100)
        #clf = svm.SVC(kernel='rbf', C=1).fit(X_train, y_train)
        clf = svm.SVC(kernel='linear', C=1).fit(X_train, y_train)
        
        #clf = clf.fit(X_train, y_train)
        score=clf.score(X_test, y_test)   
        result[m.columns[i]]=score
        print m.columns[i],score
    return result
    
def TestDays():
    """
    1.DaysInHospital_Y20 = -0.679463423154 
    2.DaysInHospital_Y215 = 0.679839894557
    test_size=0.95
    1->0.849336540016
    1+2->0.86902890096
    all->0.906261684227
    all+16->0.906261684227
    all+paydelay76->0.906261684227
    
    test_size=0.9
    all->0.906158873286
    
    """
    from sklearn import cross_validation
    from sklearn import svm
    import itertools
    from sklearn.ensemble import RandomForestClassifier
    #itertools.permutations(["DaysInHospital_Y20","DaysInHospital_Y215","DaysInHospital_Y212","DaysInHospital_Y26","DaysInHospital_Y28","DaysInHospital_Y210", "DaysInHospital_Y214"])
    features=[
     "DaysInHospital_Y26",
     "DaysInHospital_Y212",
     "DaysInHospital_Y214",
     "DaysInHospital_Y210",
     "DaysInHospital_Y28",
     "DaysInHospital_Y20",
     "DaysInHospital_Y215" 
    ]
    combs = []
    for i in xrange(1, len(features)+1):
        #combs.append(i)
        els = [list(x) for x in itertools.combinations(features, i)]
        combs.extend(els)
    #return combs    
    m=pd.load("featuresSet")
    result={}    
    for i in combs:
        X=m[i]
        Y=m["DaysInHospital_Y2"]
        X_train, X_test, y_train, y_test=cross_validation.train_test_split(X,Y,test_size=0.95,random_state=0)
        #clf = RandomForestClassifier(n_estimators=100)
        #clf = svm.SVC(kernel='rbf', C=1).fit(X_train, y_train)
        clf = svm.SVC(kernel='linear', C=1).fit(X_train, y_train)
        
        #clf = clf.fit(X_train, y_train)
        score=clf.score(X_test, y_test)   
        print i,"=>",score
        result[str(i)]=score
    return result


def Merge():
    dihY2   =pd.read_csv("DayInHospital_Y2.csv")
    dihY2.rename(columns={'memberid':'MemberID'},inplace=True)
    claims=pd.read_csv("Claims_Y1.csv")        
    m=pd.merge(claims,dihY2,on='MemberID')
    member=pd.read_csv("Members_Y1.csv")
    m=pd.merge(m,member,on='MemberID')    
    return m
    #expend features
    """
    Index([u'MemberID', u'ProviderID', u'vendor', u'pcp', u'Year', u'specialty', u'placesvc', u'paydelay', u'LengthOfStay', u'dsfs',
    u'PrimaryConditionGroup', u'CharlsonIndex', u'DaysInHospital_Y2', u'sex', u'AgeAtFirstClaim'], dtype='object')
    MemberID (77289,)
    ProviderID (7825,)
    vendor (3851,)
    pcp (1012,)
    Year (1,)
    specialty (12,)
    placesvc (8,)
    paydelay (163,)
    LengthOfStay (13,)
    dsfs (12,)
    PrimaryConditionGroup (45,)
    CharlsonIndex (4,)
    DaysInHospital_Y2 (16,)
    sex (2,)
    AgeAtFirstClaim (9,)
    
    new features:15-299
    """
    for i in m.columns:
        
        if i=="MemberID" or i=="ProviderID" or i=="vendor" or i=="pcp" or i=="Year":
            continue
        
        else:
            categories=m[i].unique()
            for j in categories:
                #print j
                print m.shape
                fieldName=i+str(j)
                fieldName=fieldName.replace(" ","_")
                fieldName=fieldName.replace(".0","")                            
                m[fieldName]=pd.Series((m[i]==j)+0, index=m.index)
                #print m[fieldName]
                #raw_input("wait for me")
            
    return m
    
    
"""
{'AgeAtFirstClaim0-9': -0.051990535358995144,
 'AgeAtFirstClaim10-19': -0.053120585214221083,
 'AgeAtFirstClaim20-29': -0.020055519475893367,
 'AgeAtFirstClaim30-39': -0.034121723301183973,
 'AgeAtFirstClaim40-49': -0.070829706576915008,
 'AgeAtFirstClaim50-59': -0.051728956041176576,
 'AgeAtFirstClaim60-69': -0.012948134422449459,
 'AgeAtFirstClaim70-79': 0.064311863282610862,
 'AgeAtFirstClaim80+': 0.13442560590738883,
 'CharlsonIndex0': -0.13824233051485832,
 'CharlsonIndex1-2': 0.12582855508262986,
 'CharlsonIndex3-4': 0.047233068655335698,
 'CharlsonIndex5+': 0.010936434562424569,
 'DaysInHospital_Y20': -0.67946342315363084,
 'DaysInHospital_Y21': -0.0087339388361676908,
 'DaysInHospital_Y210': 0.23731146168158038,
 'DaysInHospital_Y211': 0.1710723543155184,
 'DaysInHospital_Y212': 0.236856544714515,
 'DaysInHospital_Y213': 0.16892919181367275,
 'DaysInHospital_Y214': 0.23558385967427445,
 'DaysInHospital_Y215': 0.67983989455717353,
 'DaysInHospital_Y22': 0.050406664146141519,
 'DaysInHospital_Y23': 0.079268960824596391,
 'DaysInHospital_Y24': 0.17473576356507209,
 'DaysInHospital_Y25': 0.13826840457010925,
 'DaysInHospital_Y26': 0.20653378434839736,
 'DaysInHospital_Y27': 0.16362134058527042,
 'DaysInHospital_Y28': 0.23918670396271141,
 'DaysInHospital_Y29': 0.17948641033261789,
 'LengthOfStay1-_2_weeks': 0.016002714471196423,
 'LengthOfStay12-26_weeks': -0.0005386616908135206,
 'LengthOfStay1_day': 0.017734250102598718,
 'LengthOfStay2-_4_weeks': 0.011378504058779076,
 'LengthOfStay26+_weeks': 0.0018575051880313092,
 'LengthOfStay2_days': 0.010544670413215319,
 'LengthOfStay3_days': 0.003834178163569857,
 'LengthOfStay4-_8_weeks': 0.015374103564451316,
 'LengthOfStay4_days': 0.0029952229839018997,
 'LengthOfStay5_days': 0.005178171758455947,
 'LengthOfStay6_days': 0.00097592601611427572,
 'LengthOfStay8-12_weeks': 0.0037639000845775205,
 #'LengthOfStaynan': nan,
 'PrimaryConditionGroupAMI': 0.030610758769220529,
 'PrimaryConditionGroupAPPCHOL': 0.0012128879780080107,
 'PrimaryConditionGroupARTHSPIN': -0.0035520633979465302,
 'PrimaryConditionGroupCANCRA': 0.0035063605670472044,
 'PrimaryConditionGroupCANCRB': 0.01046496422568827,
 'PrimaryConditionGroupCANCRM': 0.0039967413725477026,
 'PrimaryConditionGroupCATAST': 0.0085184472783841995,
 'PrimaryConditionGroupCHF': 0.030577436050487891,
 'PrimaryConditionGroupCOPD': 0.016411282973425607,
 'PrimaryConditionGroupFLaELEC': 0.0073563987878677017,
 'PrimaryConditionGroupFXDISLC': -0.010069047003729216,
 'PrimaryConditionGroupGIBLEED': 0.018035041532077942,
 'PrimaryConditionGroupGIOBSENT': -0.0014232966616945406,
 'PrimaryConditionGroupGYNEC1': -0.014752923195988022,
 'PrimaryConditionGroupGYNECA': -0.0050279477676819723,
 'PrimaryConditionGroupHEART2': 0.032848460309481527,
 'PrimaryConditionGroupHEART4': 0.011181089744136057,
 'PrimaryConditionGroupHEMTOL': 0.019245959221376487,
 'PrimaryConditionGroupHIPFX': 0.010136020107810244,
 'PrimaryConditionGroupINFEC4': -0.014684059391852832,
 'PrimaryConditionGroupLIVERDZ': -0.00084052826229429262,
 'PrimaryConditionGroupMETAB1': 0.0025547248267555832,
 'PrimaryConditionGroupMETAB3': -0.0057154033867015596,
 'PrimaryConditionGroupMISCHRT': 0.0032039733347676439,
 'PrimaryConditionGroupMISCL1': 0.0008906653902656513,
 'PrimaryConditionGroupMISCL5': 0.00024306204742741011,
 'PrimaryConditionGroupMSC2a3': -0.032360877340658872,
 'PrimaryConditionGroupNEUMENT': -0.0015903552139161674,
 'PrimaryConditionGroupODaBNCA': -0.012518443796833612,
 'PrimaryConditionGroupPERINTL': -0.0015112797623188408,
 'PrimaryConditionGroupPERVALV': 0.0022840476904212627,
 'PrimaryConditionGroupPNCRDZ': 0.00050839345982758529,
 'PrimaryConditionGroupPNEUM': 0.0068629547626261724,
 'PrimaryConditionGroupPRGNCY': 0.0059664273791333955,
 'PrimaryConditionGroupRENAL1': 0.0050715918738550936,
 'PrimaryConditionGroupRENAL2': 0.034338474947324282,
 'PrimaryConditionGroupRENAL3': 0.0025755303445003474,
 'PrimaryConditionGroupRESPR4': -0.017525071823173366,
 'PrimaryConditionGroupROAMI': 0.025019415660643404,
 'PrimaryConditionGroupSEIZURE': 0.011789752443471919,
 'PrimaryConditionGroupSEPSIS': 0.00399247371507927,
 'PrimaryConditionGroupSKNAUT': -0.007385328441171464,
 'PrimaryConditionGroupSTROKE': 0.011273321660608532,
 'PrimaryConditionGroupTRAUMA': -0.013109609509598077,
 'PrimaryConditionGroupUTI': -0.0013519147971185026,
 'dsfs0-_1_month': -0.057138676661269094,
 'dsfs1-_2_months': -0.0010176712515487575,
 'dsfs10-11_months': 0.022174259224654772,
 'dsfs11-12_months': 0.021351925291324389,
 'dsfs2-_3_months': 0.0034856194173231679,
 'dsfs3-_4_months': 0.0068630200084156025,
 'dsfs4-_5_months': 0.0092195808845035414,
 'dsfs5-_6_months': 0.0088008967415616896,
 'dsfs6-_7_months': 0.009868797199045461,
 'dsfs7-_8_months': 0.0089013577662562116,
 'dsfs8-_9_months': 0.010949392053089547,
 'dsfs9-10_months': 0.015966734567246895,
 'paydelay0': 0.0010521142646558442,
 'paydelay1': -0.00092969779605539191,
 'paydelay10': -0.0025502610280857766,
 'paydelay100': -0.0021948827214483672,
 'paydelay101': -0.0022543582507939283,
 'paydelay102': -0.00076754526042429954,
 'paydelay103': -0.000271468074138099,
 'paydelay104': 0.0014270951644715379,
 'paydelay105': 0.00026535315176085108,
 'paydelay106': -0.00030860347235817781,
 'paydelay107': 0.00057344716599231282,
 'paydelay108': -0.00043698867854749383,
 'paydelay109': 0.00068037487483748205,
 'paydelay11': 0.00037663254145464274,
 'paydelay110': -0.0010481655110123355,
 'paydelay111': -0.0004457543736642376,
 'paydelay112': -0.0023746402996821588,
 'paydelay113': -0.00090301425209026679,
 'paydelay114': -0.00076817015232317446,
 'paydelay115': -0.0010761821218718019,
 'paydelay116': 0.0017048715221576682,
 'paydelay117': 0.0038798220390708852,
 'paydelay118': 0.0017741585359746747,
 'paydelay119': 0.00028431984604136999,
 'paydelay12': -0.0020368549866071461,
 'paydelay120': -0.0036680198223440121,
 'paydelay121': -0.00065209103856827687,
 'paydelay122': 0.0012017206522238117,
 'paydelay123': 2.5550710863681468e-05,
 'paydelay124': 0.0008210810783944515,
 'paydelay125': 0.0032403904494617826,
 'paydelay126': 7.4339891084732973e-05,
 'paydelay127': 0.0024288760160666866,
 'paydelay128': 0.0020189089646596215,
 'paydelay129': -0.0013200063738423974,
 'paydelay13': -0.001636357434912249,
 'paydelay130': -0.000842517470032419,
 'paydelay131': 0.0020888212601223913,
 'paydelay132': 0.00099817479090601874,
 'paydelay133': 0.0014182099099480205,
 'paydelay134': -0.0011503125677089919,
 'paydelay135': -0.00083604595560671203,
 'paydelay136': -0.0023341381163196392,
 'paydelay137': 0.0014618489569780187,
 'paydelay138': 0.00050695046093760378,
 'paydelay139': -0.00076701557983496619,
 'paydelay14': -0.001893447121653207,
 'paydelay140': -0.00055311865636457642,
 'paydelay141': 0.0004516991751084062,
 'paydelay142': 0.00074921851263036259,
 'paydelay143': 0.0016783255939981708,
 'paydelay144': 0.00027601465640864249,
 'paydelay145': -0.0018700317583144521,
 'paydelay146': 0.00093681953931392628,
 'paydelay147': 0.0018947403226469138,
 'paydelay148': -0.0015966683779331588,
 'paydelay149': 0.0013937722871974827,
 'paydelay15': -0.0026777360291627781,
 'paydelay150': -0.00018838730092220093,
 'paydelay151': -0.0018954405801069163,
 'paydelay152': -0.00040537224202083261,
 'paydelay153': 0.00027323334292028072,
 'paydelay154': -0.00060902775054555425,
 'paydelay155': 1.5935013123199077e-05,
 'paydelay156': -0.0023379371073389975,
 'paydelay157': 0.0012608020158976373,
 'paydelay158': 0.00062167705798619403,
 'paydelay159': 0.0016857095281476435,
 'paydelay16': -0.0033853331634081251,
 'paydelay160': -0.0012190988503349804,
 'paydelay161': -0.0011328509624568633,
 'paydelay17': -0.0042140380542248076,
 'paydelay18': -0.0058557031557985228,
 'paydelay19': -0.0030912553793221312,
 'paydelay2': 0.0014038357567363574,
 'paydelay20': -0.0023940881501139616,
 'paydelay21': -0.0051321873247753662,
 'paydelay22': 0.00049039358967818865,
 'paydelay23': 0.00086120951685571886,
 'paydelay24': 0.0016050619844855205,
 'paydelay25': -0.003331019095847407,
 'paydelay26': 0.00085687351416086565,
 'paydelay27': 0.00023953372171437996,
 'paydelay28': 0.00018449120684821204,
 'paydelay29': 0.001819012213016514,
 'paydelay3': -0.00071354694818117237,
 'paydelay30': -0.00014423920271112382,
 'paydelay31': 0.0016133046090605315,
 'paydelay32': -0.00022755699537699975,
 'paydelay33': 0.0019784784514850682,
 'paydelay34': 0.00040667101071007508,
 'paydelay35': -0.0003943430495734369,
 'paydelay36': 0.0020434853382608996,
 'paydelay37': 0.0028807577930185951,
 'paydelay38': -0.00029577873301640806,
 'paydelay39': -0.00099496510649477003,
 'paydelay4': -0.0016567159620019646,
 'paydelay40': 0.0046244091154052855,
 'paydelay41': 0.0017429132865842193,
 'paydelay42': -0.0014398584785627537,
 'paydelay43': 0.00031796050710647188,
 'paydelay44': 0.00069441307348550919,
 'paydelay45': 0.0015827338196245828,
 'paydelay46': 0.0011842733231384802,
 'paydelay47': 0.0094826817560977449,
 'paydelay48': 0.0038598870396361995,
 'paydelay49': 0.0025892810697994884,
 'paydelay5': -0.00095716436296253355,
 'paydelay50': 0.0013784816139350302,
 'paydelay51': 0.0052888330038332373,
 'paydelay52': 0.0043719061139536365,
 'paydelay53': 0.00076936561131082882,
 'paydelay54': 0.0030249319260404715,
 'paydelay55': 0.0037535297572629913,
 'paydelay56': -0.00037001726288951456,
 'paydelay57': 0.00094027381845838032,
 'paydelay58': 0.00091308544268382948,
 'paydelay59': -0.0023924908068606713,
 'paydelay6': -0.001077564540917901,
 'paydelay60': -0.0019681620255400196,
 'paydelay61': 0.0031032076237671475,
 'paydelay62': -7.0868717864139321e-05,
 'paydelay63': -0.003745593432843248,
 'paydelay64': 6.3218315869063629e-05,
 'paydelay65': 0.001953719689020343,
 'paydelay66': -0.0028456409906599104,
 'paydelay67': -0.0035566623552632736,
 'paydelay68': 0.0013236523155071083,
 'paydelay69': 0.00022040552408937974,
 'paydelay7': -0.00032590406651093306,
 'paydelay70': -0.0038638803734610209,
 'paydelay71': -0.0033764190361944132,
 'paydelay72': -0.0038707936862452546,
 'paydelay73': -0.0032796771173344481,
 'paydelay74': -0.00091994208203768587,
 'paydelay75': -0.00010530661341791988,
 'paydelay76': 3.6871601203532467e-05,
 'paydelay77': -0.0012384775983214993,
 'paydelay78': -0.0011052492004519157,
 'paydelay79': -0.004198329894719203,
 'paydelay8': -0.0019500907711411293,
 'paydelay80': -0.0012343309007495648,
 'paydelay81': -0.0017991867375654399,
 'paydelay82': -0.0010852277906497831,
 'paydelay83': 0.00014023613100656252,
 'paydelay84': -0.0030434918637921781,
 'paydelay85': -0.00065015308895255647,
 'paydelay86': -0.0025513891738748111,
 'paydelay87': 0.00065330255159183777,
 'paydelay88': -0.001710048085560648,
 'paydelay89': 0.0011907776389895187,
 'paydelay9': -0.00041272688407094002,
 'paydelay90': -0.00066719121993306972,
 'paydelay91': -0.00048590370692681736,
 'paydelay92': -4.2848067783431625e-05,
 'paydelay93': -0.00099883873192105062,
 'paydelay94': -0.0013828011249261995,
 'paydelay95': -0.00083475409542182663,
 'paydelay96': -2.2257659964384273e-05,
 'paydelay97': 0.00078055295242069537,
 'paydelay98': 0.0009529945677058465,
 'paydelay99': 0.0009572416524036379,
 #'paydelaynan': nan,
 'placesvcAmbulance': -0.0088951291144559845,
 'placesvcHome': 0.022999478492210988,
 'placesvcIndependent_Lab': -0.0091750319900114063,
 'placesvcInpatient_Hospital': 0.067024686882097659,
 'placesvcOffice': -0.041390493806630409,
 'placesvcOther': 0.032556303510249691,
 'placesvcOutpatient_Hospital': 0.0073223258311093725,
 'placesvcUrgent_Care': 0.02925019299131448,
 'sexF': 0.0058870678746491449,
 'sexM': -0.0058870678746491207,
 'specialtyAnesthesiology': 0.00067542493953508763,
 'specialtyDiagnostic_Imaging': 0.013012588534897708,
 'specialtyEmergency': 0.018652276647575378,
 'specialtyGeneral_Practice': -0.03317219944350918,
 'specialtyInternal': 0.031778661136686516,
 'specialtyLaboratory': -0.0082867311504348142,
 'specialtyObstetrics_and_Gynecology': -0.01497379196138463,
 'specialtyOther': 0.014017596989319909,
 'specialtyPathology': -0.0028972566559410341,
 'specialtyPediatrics': -0.041589311514704998,
 'specialtyRehabilitation': -0.0042221748355297222,
 'specialtySurgery': 0.0047687295186776306}

===================sort ==============================
[('DaysInHospital_Y20', -0.6794634231536308),
 ('CharlsonIndex0', -0.13824233051485832),
 ('AgeAtFirstClaim40-49', -0.07082970657691501),
 ('dsfs0-_1_month', -0.057138676661269094),
 ('AgeAtFirstClaim10-19', -0.05312058521422108),
 ('AgeAtFirstClaim0-9', -0.051990535358995144),
 ('AgeAtFirstClaim50-59', -0.051728956041176576),
 ('specialtyPediatrics', -0.041589311514705),
 ('placesvcOffice', -0.04139049380663041),
 ('AgeAtFirstClaim30-39', -0.03412172330118397),
 ('specialtyGeneral_Practice', -0.03317219944350918),
 ('PrimaryConditionGroupMSC2a3', -0.03236087734065887),
 ('AgeAtFirstClaim20-29', -0.020055519475893367),
 ('PrimaryConditionGroupRESPR4', -0.017525071823173366),
 ('specialtyObstetrics_and_Gynecology', -0.01497379196138463),
 ('PrimaryConditionGroupGYNEC1', -0.014752923195988022),
 ('PrimaryConditionGroupINFEC4', -0.014684059391852832),
 ('PrimaryConditionGroupTRAUMA', -0.013109609509598077),
 ('AgeAtFirstClaim60-69', -0.012948134422449459),
 ('PrimaryConditionGroupODaBNCA', -0.012518443796833612),
 ('PrimaryConditionGroupFXDISLC', -0.010069047003729216),
 ('placesvcIndependent_Lab', -0.009175031990011406),
 ('placesvcAmbulance', -0.008895129114455985),
 ('DaysInHospital_Y21', -0.00873393883616769),
 ('specialtyLaboratory', -0.008286731150434814),
 ('PrimaryConditionGroupSKNAUT', -0.007385328441171464),
 ('sexM', -0.005887067874649121),
 ('paydelay18', -0.005855703155798523),
 ('PrimaryConditionGroupMETAB3', -0.00571540338670156),
 ('paydelay21', -0.005132187324775366),
 ('PrimaryConditionGroupGYNECA', -0.005027947767681972),
 ('specialtyRehabilitation', -0.004222174835529722),
 ('paydelay17', -0.004214038054224808),
 ('paydelay79', -0.004198329894719203),
 ('paydelay72', -0.0038707936862452546),
 ('paydelay70', -0.003863880373461021),
 ('paydelay63', -0.003745593432843248),
 ('paydelay120', -0.003668019822344012),
 ('paydelay67', -0.0035566623552632736),
 ('PrimaryConditionGroupARTHSPIN', -0.00355206339794653),
 ('paydelay16', -0.003385333163408125),
 ('paydelay71', -0.003376419036194413),
 ('paydelay25', -0.003331019095847407),
 ('paydelay73', -0.003279677117334448),
 ('paydelay19', -0.003091255379322131),
 ('paydelay84', -0.003043491863792178),
 ('specialtyPathology', -0.002897256655941034),
 ('paydelay66', -0.0028456409906599104),
 ('paydelay15', -0.002677736029162778),
 ('paydelay86', -0.002551389173874811),
 ('paydelay10', -0.0025502610280857766),
 ('paydelay20', -0.0023940881501139616),
 ('paydelay59', -0.0023924908068606713),
 ('paydelay112', -0.0023746402996821588),
 ('paydelay156', -0.0023379371073389975),
 ('paydelay136', -0.002334138116319639),
 ('paydelay101', -0.0022543582507939283),
 ('paydelay100', -0.002194882721448367),
 ('paydelay12', -0.002036854986607146),
 ('paydelay60', -0.0019681620255400196),
 ('paydelay8', -0.0019500907711411293),
 ('paydelay151', -0.0018954405801069163),
 ('paydelay14', -0.001893447121653207),
 ('paydelay145', -0.0018700317583144521),
 ('paydelay81', -0.0017991867375654399),
 ('paydelay88', -0.001710048085560648),
 ('paydelay4', -0.0016567159620019646),
 ('paydelay13', -0.001636357434912249),
 ('paydelay148', -0.0015966683779331588),
 ('PrimaryConditionGroupNEUMENT', -0.0015903552139161674),
 ('PrimaryConditionGroupPERINTL', -0.0015112797623188408),
 ('paydelay42', -0.0014398584785627537),
 ('PrimaryConditionGroupGIOBSENT', -0.0014232966616945406),
 ('paydelay94', -0.0013828011249261995),
 ('PrimaryConditionGroupUTI', -0.0013519147971185026),
 ('paydelay129', -0.0013200063738423974),
 ('paydelay77', -0.0012384775983214993),
 ('paydelay80', -0.0012343309007495648),
 ('paydelay160', -0.0012190988503349804),
 ('paydelay134', -0.001150312567708992),
 ('paydelay161', -0.0011328509624568633),
 ('paydelay78', -0.0011052492004519157),
 ('paydelay82', -0.001085227790649783),
 ('paydelay6', -0.001077564540917901),
 ('paydelay115', -0.001076182121871802),
 ('paydelay110', -0.0010481655110123355),
 ('dsfs1-_2_months', -0.0010176712515487575),
 ('paydelay93', -0.0009988387319210506),
 ('paydelay39', -0.00099496510649477),
 ('paydelay5', -0.0009571643629625335),
 ('paydelay1', -0.0009296977960553919),
 ('paydelay74', -0.0009199420820376859),
 ('paydelay113', -0.0009030142520902668),
 ('paydelay130', -0.000842517470032419),
 ('PrimaryConditionGroupLIVERDZ', -0.0008405282622942926),
 ('paydelay135', -0.000836045955606712),
 ('paydelay95', -0.0008347540954218266),
 ('paydelay114', -0.0007681701523231745),
 ('paydelay102', -0.0007675452604242995),
 ('paydelay139', -0.0007670155798349662),
 ('paydelay3', -0.0007135469481811724),
 ('paydelay90', -0.0006671912199330697),
 ('paydelay121', -0.0006520910385682769),
 ('paydelay85', -0.0006501530889525565),
 ('paydelay154', -0.0006090277505455542),
 ('paydelay140', -0.0005531186563645764),
 ('LengthOfStay12-26_weeks', -0.0005386616908135206),
 ('paydelay91', -0.00048590370692681736),
 ('paydelay111', -0.0004457543736642376),
 ('paydelay108', -0.0004369886785474938),
 ('paydelay9', -0.00041272688407094),
 ('paydelay152', -0.0004053722420208326),
 ('paydelay35', -0.0003943430495734369),
 ('paydelay56', -0.00037001726288951456),
 ('paydelay7', -0.00032590406651093306),
 ('paydelay106', -0.0003086034723581778),
 ('paydelay38', -0.00029577873301640806),
 ('paydelay103', -0.000271468074138099),
 ('paydelay32', -0.00022755699537699975),
 ('paydelay150', -0.00018838730092220093),
 ('paydelay30', -0.00014423920271112382),
 ('paydelay75', -0.00010530661341791988),
 ('paydelay62', -7.086871786413932e-05),
 ('paydelay92', -4.2848067783431625e-05),
 ('paydelay96', -2.2257659964384273e-05),
 ('paydelay155', 1.5935013123199077e-05),
 ('paydelay123', 2.5550710863681468e-05),
 ('paydelay76', 3.687160120353247e-05),
 ('paydelay64', 6.321831586906363e-05),
 ('paydelay126', 7.433989108473297e-05),
 ('paydelay83', 0.00014023613100656252),
 ('paydelay28', 0.00018449120684821204),
 ('paydelay69', 0.00022040552408937974),
 ('paydelay27', 0.00023953372171437996),
 ('PrimaryConditionGroupMISCL5', 0.00024306204742741011),
 ('paydelay105', 0.0002653531517608511),
 ('paydelay153', 0.0002732333429202807),
 ('paydelay144', 0.0002760146564086425),
 ('paydelay119', 0.00028431984604137),
 ('paydelay43', 0.0003179605071064719),
 ('paydelay11', 0.00037663254145464274),
 ('paydelay34', 0.0004066710107100751),
 ('paydelay141', 0.0004516991751084062),
 ('paydelay22', 0.0004903935896781886),
 ('paydelay138', 0.0005069504609376038),
 ('PrimaryConditionGroupPNCRDZ', 0.0005083934598275853),
 ('paydelay107', 0.0005734471659923128),
 ('paydelay158', 0.000621677057986194),
 ('paydelay87', 0.0006533025515918378),
 ('specialtyAnesthesiology', 0.0006754249395350876),
 ('paydelay109', 0.000680374874837482),
 ('paydelay44', 0.0006944130734855092),
 ('paydelay142', 0.0007492185126303626),
 ('paydelay53', 0.0007693656113108288),
 ('paydelay97', 0.0007805529524206954),
 ('paydelay124', 0.0008210810783944515),
 ('paydelay26', 0.0008568735141608657),
 ('paydelay23', 0.0008612095168557189),
 ('PrimaryConditionGroupMISCL1', 0.0008906653902656513),
 ('paydelay58', 0.0009130854426838295),
 ('paydelay146', 0.0009368195393139263),
 ('paydelay57', 0.0009402738184583803),
 ('paydelay98', 0.0009529945677058465),
 ('paydelay99', 0.0009572416524036379),
 ('LengthOfStay6_days', 0.0009759260161142757),
 ('paydelay132', 0.0009981747909060187),
 ('paydelay0', 0.0010521142646558442),
 ('paydelay46', 0.0011842733231384802),
 ('paydelay89', 0.0011907776389895187),
 ('paydelay122', 0.0012017206522238117),
 ('PrimaryConditionGroupAPPCHOL', 0.0012128879780080107),
 ('paydelay157', 0.0012608020158976373),
 ('paydelay68', 0.0013236523155071083),
 ('paydelay50', 0.0013784816139350302),
 ('paydelay149', 0.0013937722871974827),
 ('paydelay2', 0.0014038357567363574),
 ('paydelay133', 0.0014182099099480205),
 ('paydelay104', 0.001427095164471538),
 ('paydelay137', 0.0014618489569780187),
 ('paydelay45', 0.0015827338196245828),
 ('paydelay24', 0.0016050619844855205),
 ('paydelay31', 0.0016133046090605315),
 ('paydelay143', 0.0016783255939981708),
 ('paydelay159', 0.0016857095281476435),
 ('paydelay116', 0.0017048715221576682),
 ('paydelay41', 0.0017429132865842193),
 ('paydelay118', 0.0017741585359746747),
 ('paydelay29', 0.001819012213016514),
 ('LengthOfStay26+_weeks', 0.0018575051880313092),
 ('paydelay147', 0.0018947403226469138),
 ('paydelay65', 0.001953719689020343),
 ('paydelay33', 0.001978478451485068),
 ('paydelay128', 0.0020189089646596215),
 ('paydelay36', 0.0020434853382608996),
 ('paydelay131', 0.0020888212601223913),
 ('PrimaryConditionGroupPERVALV', 0.0022840476904212627),
 ('paydelay127', 0.0024288760160666866),
 ('PrimaryConditionGroupMETAB1', 0.0025547248267555832),
 ('PrimaryConditionGroupRENAL3', 0.0025755303445003474),
 ('paydelay49', 0.0025892810697994884),
 ('paydelay37', 0.002880757793018595),
 ('LengthOfStay4_days', 0.0029952229839018997),
 ('paydelay54', 0.0030249319260404715),
 ('paydelay61', 0.0031032076237671475),
 ('PrimaryConditionGroupMISCHRT', 0.003203973334767644),
 ('paydelay125', 0.0032403904494617826),
 ('dsfs2-_3_months', 0.003485619417323168),
 ('PrimaryConditionGroupCANCRA', 0.0035063605670472044),
 ('paydelay55', 0.0037535297572629913),
 ('LengthOfStay8-12_weeks', 0.0037639000845775205),
 ('LengthOfStay3_days', 0.003834178163569857),
 ('paydelay48', 0.0038598870396361995),
 ('paydelay117', 0.003879822039070885),
 ('PrimaryConditionGroupSEPSIS', 0.00399247371507927),
 ('PrimaryConditionGroupCANCRM', 0.003996741372547703),
 ('paydelay52', 0.0043719061139536365),
 ('paydelay40', 0.0046244091154052855),
 ('specialtySurgery', 0.0047687295186776306),
 ('PrimaryConditionGroupRENAL1', 0.005071591873855094),
 ('LengthOfStay5_days', 0.005178171758455947),
 ('paydelay51', 0.005288833003833237),
 ('sexF', 0.005887067874649145),
 ('PrimaryConditionGroupPRGNCY', 0.0059664273791333955),
 ('PrimaryConditionGroupPNEUM', 0.006862954762626172),
 ('dsfs3-_4_months', 0.0068630200084156025),
 ('placesvcOutpatient_Hospital', 0.0073223258311093725),
 ('PrimaryConditionGroupFLaELEC', 0.007356398787867702),
 ('PrimaryConditionGroupCATAST', 0.0085184472783842),
 ('dsfs5-_6_months', 0.00880089674156169),
 ('dsfs7-_8_months', 0.008901357766256212),
 ('dsfs4-_5_months', 0.009219580884503541),
 ('paydelay47', 0.009482681756097745),
 ('dsfs6-_7_months', 0.009868797199045461),
 ('PrimaryConditionGroupHIPFX', 0.010136020107810244),
 ('PrimaryConditionGroupCANCRB', 0.01046496422568827),
 ('LengthOfStay2_days', 0.010544670413215319),
 ('CharlsonIndex5+', 0.010936434562424569),
 ('dsfs8-_9_months', 0.010949392053089547),
 ('PrimaryConditionGroupHEART4', 0.011181089744136057),
 ('PrimaryConditionGroupSTROKE', 0.011273321660608532),
 ('LengthOfStay2-_4_weeks', 0.011378504058779076),
 ('PrimaryConditionGroupSEIZURE', 0.01178975244347192),
 ('specialtyDiagnostic_Imaging', 0.013012588534897708),
 ('specialtyOther', 0.01401759698931991),
 ('LengthOfStay4-_8_weeks', 0.015374103564451316),
 ('dsfs9-10_months', 0.015966734567246895),
 ('LengthOfStay1-_2_weeks', 0.016002714471196423),
 ('PrimaryConditionGroupCOPD', 0.016411282973425607),
 ('LengthOfStay1_day', 0.017734250102598718),
 ('PrimaryConditionGroupGIBLEED', 0.018035041532077942),
 ('specialtyEmergency', 0.018652276647575378),
 ('PrimaryConditionGroupHEMTOL', 0.019245959221376487),
 ('dsfs11-12_months', 0.02135192529132439),
 ('dsfs10-11_months', 0.022174259224654772),
 ('placesvcHome', 0.022999478492210988),
 ('PrimaryConditionGroupROAMI', 0.025019415660643404),
 ('placesvcUrgent_Care', 0.02925019299131448),
 ('PrimaryConditionGroupCHF', 0.03057743605048789),
 ('PrimaryConditionGroupAMI', 0.03061075876922053),
 ('specialtyInternal', 0.031778661136686516),
 ('placesvcOther', 0.03255630351024969),
 ('PrimaryConditionGroupHEART2', 0.03284846030948153),
 ('PrimaryConditionGroupRENAL2', 0.03433847494732428),
 ('CharlsonIndex3-4', 0.0472330686553357),
 ('DaysInHospital_Y22', 0.05040666414614152),
 ('AgeAtFirstClaim70-79', 0.06431186328261086),
 ('placesvcInpatient_Hospital', 0.06702468688209766),
 ('DaysInHospital_Y23', 0.07926896082459639),
 ('CharlsonIndex1-2', 0.12582855508262986),
 ('AgeAtFirstClaim80+', 0.13442560590738883),
 ('DaysInHospital_Y25', 0.13826840457010925),
 ('DaysInHospital_Y27', 0.16362134058527042),
 ('DaysInHospital_Y213', 0.16892919181367275),
 ('DaysInHospital_Y211', 0.1710723543155184),
 ('DaysInHospital_Y24', 0.1747357635650721),
 ('DaysInHospital_Y29', 0.1794864103326179),
 ('DaysInHospital_Y26', 0.20653378434839736),
 ('DaysInHospital_Y214', 0.23558385967427445),
 ('DaysInHospital_Y212', 0.236856544714515),
 ('DaysInHospital_Y210', 0.23731146168158038),
 ('DaysInHospital_Y28', 0.2391867039627114),
 ('DaysInHospital_Y215', 0.6798398945571735)]



"""
def TestDays2():
    """
    ['AgeAtFirstClaim80+'] => 0.76758899605
    ['CharlsonIndex1-2'] => 0.76758899605
    ['CharlsonIndex0'] => 0.76758899605
    ['AgeAtFirstClaim80+', 'CharlsonIndex1-2'] => 0.76758899605
    ['AgeAtFirstClaim80+', 'CharlsonIndex0'] => 0.76758899605
    ['CharlsonIndex1-2', 'CharlsonIndex0'] => 0.76758899605
    ['AgeAtFirstClaim80+', 'CharlsonIndex1-2', 'CharlsonIndex0'] => 0.76758899605    
    """
    from sklearn import cross_validation
    from sklearn import svm
    import itertools
    from sklearn.ensemble import RandomForestClassifier
    #itertools.permutations(["DaysInHospital_Y20","DaysInHospital_Y215","DaysInHospital_Y212","DaysInHospital_Y26","DaysInHospital_Y28","DaysInHospital_Y210", "DaysInHospital_Y214"])
    m=pd.read_pickle("featuresSet")
    features= list(m.columns)
    """
    features=[
     "AgeAtFirstClaim80+",
     "CharlsonIndex1-2",
     "CharlsonIndex0",
    ]
    """
    
    combs = []
    for i in xrange(10,12): #xrange(100, len(features)+1):
        #combs.append(i)
        els = [list(x) for x in itertools.combinations(features, i)]
        combs.extend(els)
    #return combs
    print "features finsihed"
    result={}    
    for i in combs:
        X=m[i]
        Y=m["DaysInHospital_Y2"]
        X_train, X_test, y_train, y_test=cross_validation.train_test_split(X,Y,test_size=0.95,random_state=0)
        #clf = RandomForestClassifier(n_estimators=100)
        #clf = svm.SVC(kernel='rbf', C=1).fit(X_train, y_train)
        #clf = svm.SVC(kernel='linear', C=1).fit(X_train, y_train)
        clf = svm.LinearSVC()
        clf = clf.fit(X_train, y_train)
        score=clf.score(X_test, y_test)   
        print i,"=>",score
        result[str(i)]=score
    return result   
   
"""
#dict
{"['DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.86902890096020868,
 "['DaysInHospital_Y20']": 0.84933654001577219,
 "['DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.87493122123333189,
 "['DaysInHospital_Y210', 'DaysInHospital_Y20']": 0.85523886028889529,
 "['DaysInHospital_Y210', 'DaysInHospital_Y215']": 0.79318367726798489,
 "['DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.88486312004976564,
 "['DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.86517075910532903,
 "['DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.80311557608441864,
 "['DaysInHospital_Y210', 'DaysInHospital_Y28']": 0.78342321513998214,
 "['DaysInHospital_Y210']": 0.77349131632354839,
 "['DaysInHospital_Y212', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.873001333940709,
 "['DaysInHospital_Y212', 'DaysInHospital_Y20']": 0.85330897299627251,
 "['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.87890365421383221,
 "['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y20']": 0.8592112932693956,
 "['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y215']": 0.79715611024848521,
 "['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.88883555303026596,
 "['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.86914319208582935,
 "['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.80708800906491895,
 "['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28']": 0.78739564812048246,
 "['DaysInHospital_Y212', 'DaysInHospital_Y210']": 0.77746374930404871,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.87582922293463694,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y20']": 0.85613686199020034,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.88173154320776004,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20']": 0.86203918226332343,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y215']": 0.79998399924241315,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.89166344202419379,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.87197108107975729,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.80991589805884689,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28']": 0.79022353711441029,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210']": 0.78029163829797654,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y215']": 0.79408167896928994,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.88576112175107069,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.86606876080663409,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.80401357778572369,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28']": 0.78432121684128719,
 "['DaysInHospital_Y212', 'DaysInHospital_Y214']": 0.77438931802485345,
 "['DaysInHospital_Y212', 'DaysInHospital_Y215']": 0.79125378997536211,
 "['DaysInHospital_Y212', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.88293323275714275,
 "['DaysInHospital_Y212', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.86324087181270626,
 "['DaysInHospital_Y212', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.80118568879179586,
 "['DaysInHospital_Y212', 'DaysInHospital_Y28']": 0.78149332784735925,
 "['DaysInHospital_Y212']": 0.77156142903092551,
 "['DaysInHospital_Y214', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.87185678995413662,
 "['DaysInHospital_Y214', 'DaysInHospital_Y20']": 0.85216442900970002,
 "['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.87775911022725972,
 "['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20']": 0.85806674928282323,
 "['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y215']": 0.79601156626191283,
 "['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.88769100904369347,
 "['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.86799864809925698,
 "['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.80594346507834658,
 "['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28']": 0.78625110413390997,
 "['DaysInHospital_Y214', 'DaysInHospital_Y210']": 0.77631920531747622,
 "['DaysInHospital_Y214', 'DaysInHospital_Y215']": 0.79010924598878962,
 "['DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.88178868877057037,
 "['DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.86209632782613377,
 "['DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.80004114480522348,
 "['DaysInHospital_Y214', 'DaysInHospital_Y28']": 0.78034878386078688,
 "['DaysInHospital_Y214']": 0.77041688504435313,
 "['DaysInHospital_Y215']": 0.78728135699486179,
 "['DaysInHospital_Y26', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.88362714316269664,
 "['DaysInHospital_Y26', 'DaysInHospital_Y20']": 0.86393478221826014,
 "['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.88952946343581984,
 "['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y20']": 0.86983710249138324,
 "['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y215']": 0.80778191947047284,
 "['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.89946136225225359,
 "['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.87976900130781699,
 "['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.81771381828690659,
 "['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y28']": 0.7980214573424701,
 "['DaysInHospital_Y26', 'DaysInHospital_Y210']": 0.78808955852603635,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.88759957614319696,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y20']": 0.86790721519876046,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.89350189641632016,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y20']": 0.87380953547188356,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y215']": 0.81175435245097316,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.90343379523275391,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.88374143428831731,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.82168625126740691,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28']": 0.80199389032297042,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210']": 0.79206199150653667,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.8904274651371249,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y20']": 0.87073510419268829,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.89632978541024799,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20']": 0.87663742446581139,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y215']": 0.8145822414449011,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.90626168422668174,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.88656932328224525,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.82451414026133485,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28']": 0.80482177931689824,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210']": 0.7948898805004645,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y215']": 0.80867992117177789,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.90035936395355864,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.88066700300912204,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.81861181998821164,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28']": 0.79891945904377515,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214']": 0.7889875602273414,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y215']": 0.80585203217785006,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.8975314749596307,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.87783911401519421,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.81578393099428381,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y28']": 0.79609157004984721,
 "['DaysInHospital_Y26', 'DaysInHospital_Y212']": 0.78615967123341346,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.88645503215662458,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y20']": 0.86676267121218797,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.89235735242974767,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20']": 0.87266499148531118,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y215']": 0.81060980846440078,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.90228925124618142,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.88259689030174493,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.82054170728083453,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28']": 0.80084934633639793,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210']": 0.79091744751996418,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y215']": 0.80470748819127758,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.89638693097305833,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.87669457002862172,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.81463938700771144,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y28']": 0.79494702606327483,
 "['DaysInHospital_Y26', 'DaysInHospital_Y214']": 0.78501512724684108,
 "['DaysInHospital_Y26', 'DaysInHospital_Y215']": 0.80187959919734975,
 "['DaysInHospital_Y26', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.89355904197913039,
 "['DaysInHospital_Y26', 'DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.87386668103469389,
 "['DaysInHospital_Y26', 'DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.81181149801378349,
 "['DaysInHospital_Y26', 'DaysInHospital_Y28']": 0.792119137069347,
 "['DaysInHospital_Y26']": 0.78218723825291325,
 "['DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']": 0.87896079977664243,
 "['DaysInHospital_Y28', 'DaysInHospital_Y20']": 0.85926843883220594,
 "['DaysInHospital_Y28', 'DaysInHospital_Y215']": 0.79721325581129554,
 "['DaysInHospital_Y28']": 0.77752089486685905}
sort dict
[("['DaysInHospital_Y214']", 0.77041688504435313),
 ("['DaysInHospital_Y212']", 0.77156142903092551),
 ("['DaysInHospital_Y210']", 0.77349131632354839),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214']", 0.77438931802485345),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y210']", 0.77631920531747622),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y210']", 0.77746374930404871),
 ("['DaysInHospital_Y28']", 0.77752089486685905),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210']",
  0.78029163829797654),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y28']", 0.78034878386078688),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y28']", 0.78149332784735925),
 ("['DaysInHospital_Y26']", 0.78218723825291325),
 ("['DaysInHospital_Y210', 'DaysInHospital_Y28']", 0.78342321513998214),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28']",
  0.78432121684128719),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214']", 0.78501512724684108),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212']", 0.78615967123341346),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28']",
  0.78625110413390997),
 ("['DaysInHospital_Y215']", 0.78728135699486179),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28']",
  0.78739564812048246),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y210']", 0.78808955852603635),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214']",
  0.7889875602273414),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y215']", 0.79010924598878962),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28']",
  0.79022353711441029),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210']",
  0.79091744751996418),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y215']", 0.79125378997536211),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210']",
  0.79206199150653667),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y28']", 0.792119137069347),
 ("['DaysInHospital_Y210', 'DaysInHospital_Y215']", 0.79318367726798489),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y215']",
  0.79408167896928994),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210']",
  0.7948898805004645),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y28']",
  0.79494702606327483),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y215']",
  0.79601156626191283),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y28']",
  0.79609157004984721),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y215']",
  0.79715611024848521),
 ("['DaysInHospital_Y28', 'DaysInHospital_Y215']", 0.79721325581129554),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y28']",
  0.7980214573424701),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28']",
  0.79891945904377515),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y215']",
  0.79998399924241315),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.80004114480522348),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28']",
  0.80084934633639793),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.80118568879179586),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y215']", 0.80187959919734975),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28']",
  0.80199389032297042),
 ("['DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.80311557608441864),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.80401357778572369),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y215']",
  0.80470748819127758),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28']",
  0.80482177931689824),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y215']",
  0.80585203217785006),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.80594346507834658),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.80708800906491895),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y215']",
  0.80778191947047284),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y215']",
  0.80867992117177789),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.80991589805884689),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y215']",
  0.81060980846440078),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y215']",
  0.81175435245097316),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.81181149801378349),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y215']",
  0.8145822414449011),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.81463938700771144),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.81578393099428381),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.81771381828690659),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.81861181998821164),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.82054170728083453),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.82168625126740691),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y215']",
  0.82451414026133485),
 ("['DaysInHospital_Y20']", 0.84933654001577219),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y20']", 0.85216442900970002),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y20']", 0.85330897299627251),
 ("['DaysInHospital_Y210', 'DaysInHospital_Y20']", 0.85523886028889529),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y20']",
  0.85613686199020034),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20']",
  0.85806674928282323),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y20']",
  0.8592112932693956),
 ("['DaysInHospital_Y28', 'DaysInHospital_Y20']", 0.85926843883220594),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20']",
  0.86203918226332343),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.86209632782613377),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.86324087181270626),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y20']", 0.86393478221826014),
 ("['DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.86517075910532903),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.86606876080663409),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y20']",
  0.86676267121218797),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y20']",
  0.86790721519876046),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.86799864809925698),
 ("['DaysInHospital_Y20', 'DaysInHospital_Y215']", 0.86902890096020868),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.86914319208582935),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y20']",
  0.86983710249138324),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y20']",
  0.87073510419268829),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.87185678995413662),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.87197108107975729),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20']",
  0.87266499148531118),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.873001333940709),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y20']",
  0.87380953547188356),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.87386668103469389),
 ("['DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.87493122123333189),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.87582922293463694),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20']",
  0.87663742446581139),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.87669457002862172),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.87775911022725972),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.87783911401519421),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.87890365421383221),
 ("['DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.87896079977664243),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.87976900130781699),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.88066700300912204),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.88173154320776004),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.88178868877057037),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.88259689030174493),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.88293323275714275),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.88362714316269664),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.88374143428831731),
 ("['DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.88486312004976564),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.88576112175107069),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.88645503215662458),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20']",
  0.88656932328224525),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.88759957614319696),
 ("['DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.88769100904369347),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.88883555303026596),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.88952946343581984),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.8904274651371249),
 ("['DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.89166344202419379),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.89235735242974767),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.89350189641632016),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.89355904197913039),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.89632978541024799),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.89638693097305833),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.8975314749596307),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.89946136225225359),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.90035936395355864),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.90228925124618142),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.90343379523275391),
 ("['DaysInHospital_Y26', 'DaysInHospital_Y212', 'DaysInHospital_Y214', 'DaysInHospital_Y210', 'DaysInHospital_Y28', 'DaysInHospital_Y20', 'DaysInHospital_Y215']",
  0.90626168422668174)]



"""


def madDataTest():
    mad=pd.read_csv("../mad_mark/Data_Set1.csv")
    mad.YEAR_t.replace("Y2",2,inplace=True)
    mad.YEAR_t.replace("Y1",1,inplace=True)
    mad.YEAR_t.replace("Y3",3,inplace=True)
    #mad.YEAR_t
    #mad.DaysInHospital.corr(mad.labCount_ave)
    for i in mad.columns:
        print i,"=",mad.DaysInHospital.corr(mad[i])
    """
    MemberID_t = -0.00430663479898
    YEAR_t = -0.00927449651427
    ClaimsTruncated = 0.164786506045
    DaysInHospital = 1.0
    trainset = nan
    age_05 = -0.056000826995
    age_15 = -0.0601445986863
    age_25 = -0.00794499300771
    age_35 = -0.0307292866905
    age_45 = -0.0562487273939
    age_55 = -0.0390754357467
    age_65 = 0.00590304481323
    age_75 = 0.0710671217519
    age_85 = 0.101658629313
    age_MISS = 0.0817565359938
    sexMALE = -0.0699930216563
    sexFEMALE = -0.0389861128079
    sexMISS = 0.135082061456
    no_Claims = 0.191122215283
    no_Providers = 0.178687347617
    no_Vendors = 0.173838642771
    no_PCPs = 0.0121198329051
    no_PlaceSvcs = 0.140370771264
    no_Specialities = 0.137277180517
    no_PrimaryConditionGroups = 0.175969203523
    no_ProcedureGroups = 0.128295617971
    PayDelay_max = 0.0992427451976
    PayDelay_min = -0.0630307029362
    PayDelay_ave = -0.00278160937172
    PayDelay_stdev = 0.0608644578676
    LOS_max = 0.0797714260777
    LOS_min = 0.0667063794365
    LOS_ave = 0.0752849145203
    LOS_stdev = 0.0554133074581
    LOS_TOT_UNKNOWN = 0.186738680016
    LOS_TOT_SUPRESSED = 0.0666110899297
    LOS_TOT_KNOWN = 0.0998976373244
    dsfs_max = 0.117510833593
    dsfs_min = 0.0264104171893
    dsfs_range = 0.116861895669
    dsfs_ave = 0.115028853457
    dsfs_stdev = 0.0790663596738
    CharlsonIndexI_max = 0.152195922287
    CharlsonIndexI_min = 0.0969224672947
    CharlsonIndexI_ave = 0.128451989723
    CharlsonIndexI_range = 0.113541948358
    CharlsonIndexI_stdev = 0.0933374209541
    pcg1 = 0.0739520510236
    pcg2 = 0.0701540027474
    pcg3 = 0.073081058508
    pcg4 = 0.0598992069914
    pcg5 = 0.0206281638522
    pcg6 = 0.0631887218619
    pcg7 = 0.0358078824202
    pcg8 = 0.0802693288378
    pcg9 = 0.0184279963971
    pcg10 = 0.014268701358
    pcg11 = 0.0541839153383
    pcg12 = 0.0433781039322
    pcg13 = 0.088400528648
    pcg14 = 0.0461191112097
    pcg15 = 0.0117985040715
    pcg16 = 0.0402131926482
    pcg17 = 0.0598554861521
    pcg18 = 0.0187439344017
    pcg19 = 0.042720851816
    pcg20 = 0.00379793739162
    pcg21 = 0.0748739537197
    pcg22 = 0.0392314138858
    pcg23 = 0.0630112715716
    pcg24 = 0.0433466926681
    pcg25 = 0.044333060569
    pcg26 = 0.016335777267
    pcg27 = 0.0697714174108
    pcg28 = 0.0141534115619
    pcg29 = 0.0456076464901
    pcg30 = 0.0288611920168
    pcg31 = 0.0673984521504
    pcg32 = 0.0223942847207
    pcg33 = 0.028283477293
    pcg34 = 0.00914701124468
    pcg35 = 0.029335526754
    pcg36 = 0.0140463023075
    pcg37 = 0.0224534022296
    pcg38 = 0.0186871861388
    pcg39 = 0.0169385080861
    pcg40 = 0.0109552477521
    pcg41 = 0.0187205316449
    pcg42 = 0.00491891392811
    pcg43 = 0.0017518099961
    pcg44 = 0.00818767366843
    pcg45 = 0.0140123301906
    pcg46 = 0.00883582358705
    sp1 = 0.152888243685
    sp2 = 0.122094967867
    sp3 = 0.0640035275466
    sp4 = 0.0737214001397
    sp5 = 0.143974908375
    sp6 = 0.110255300723
    sp7 = 0.0551049348415
    sp8 = -0.0238768672022
    sp9 = 0.0186497755297
    sp10 = 0.0217599186855
    sp11 = 0.0396079256916
    sp12 = 0.0272451851112
    sp13 = 0.0310359051973
    pg1 = 0.180603203581
    pg2 = 0.130500472696
    pg3 = 0.0989434232062
    pg4 = 0.104746767113
    pg5 = 0.130613637835
    pg6 = 0.0824205260144
    pg7 = 0.0558458721711
    pg8 = 0.0494179291948
    pg9 = 0.0338332976682
    pg10 = 0.0191526309919
    pg11 = 0.0243548858649
    pg12 = 0.0299536739902
    pg13 = 0.0353286480335
    pg14 = 0.00706503580015
    pg15 = 0.021954660059
    pg16 = 0.0267342776726
    pg17 = 0.00386111888478
    pg18 = 0.0107106095828
    ps1 = 0.141355718628
    ps2 = 0.122261227779
    ps3 = 0.099203413729
    ps4 = 0.0534632445368
    ps5 = 0.11766203886
    ps6 = 0.0131791248721
    ps7 = 0.0638188178692
    ps8 = 0.0367284858509
    ps9 = 0.0257697994186
    drugCount_max = 0.144863776508
    drugCount_min = 0.0840087974342
    drugCount_ave = 0.13746106908
    drugcount_months = 0.125304891143
    labCount_max = 0.101091758852
    labCount_min = 0.0194693581889
    labCount_ave = 0.0652144810954
    labcount_months = 0.130701721624
    labNull = -0.0658304784586
    drugNull = -0.0667253925286

    """
def TestUserInY2():
    """
    第二年有住院的特性
    """
    pass

def FeatureGeneration():
    import pandas as pd
    dih=pd.read_csv("DayInHospital_Y2.csv")
    c=pd.read_csv("Claims_Y1.csv")
    dih['predict'] = pd.Series([0]*dih.shape[0], index=dih.index)
    dih['predict'] = pd.Series([0]*dih.shape[0], index=dih.index)
    dih['claims_count'] = pd.Series([0]*dih.shape[0], index=dih.index)
    dih['providers_count'] = pd.Series([0]*dih.shape[0], index=dih.index)
    dih['PCPs_count'] = pd.Series([0]*dih.shape[0], index=dih.index)
    dih['vendors_count'] = pd.Series([0]*dih.shape[0], index=dih.index)
    dih['placesvc_count'] = pd.Series([0]*dih.shape[0], index=dih.index)
    dih['specialities_count'] = pd.Series([0]*dih.shape[0], index=dih.index)
    dih['primaryconditionGroups_count'] = pd.Series([0]*dih.shape[0], index=dih.index)
    
    #dih['paydelay_count'] = pd.Series([0]*dih.shape[0], index=dih.index)
    #dih['paydelay_max'] = pd.Series([0]*dih.shape[0], index=dih.index)
    #dih['paydelay_min'] = pd.Series([0]*dih.shape[0], index=dih.index)
    #dih['paydelay_average'] = pd.Series([0]*dih.shape[0], index=dih.index)
    #dih['paydelay_var'] = pd.Series([0]*dih.shape[0], index=dih.index)
    
    
    """
    no_Claims = 0.191122215283
    no_Providers = 0.178687347617
    no_Vendors = 0.173838642771
    no_PCPs = 0.0121198329051
    no_PlaceSvcs = 0.140370771264
    no_Specialities = 0.137277180517
    no_PrimaryConditionGroups = 0.175969203523
    no_ProcedureGroups = 0.128295617971
    
    """
    total=len(dih.memberid)
    j=1
    for i in dih.memberid:    
        print j,"/",total
        j=j+1
        dih.providers_count[dih.memberid==i]=len(c[c.MemberID==i].ProviderID.unique())
        dih.PCPs_count[dih.memberid==i]=len(c[c.MemberID==i].pcp.unique())
        dih.vendors_count[dih.memberid==i]=len(c[c.MemberID==i].vendor.unique())
        dih.specialities_count[dih.memberid==i]=len(c[c.MemberID==i].specialty.unique())
        dih.primaryconditionGroups_count[dih.memberid==i]=len(c[c.MemberID==i].PrimaryConditionGroup.unique())
        dih.placesvc_count[dih.memberid==i]=len(c[c.MemberID==i].placesvc.unique())
        dih.claims_count[dih.memberid==i]=len(c[c.MemberID==i])
    return c,dih

"""
                              memberid  DaysInHospital_Y2  predict  \
memberid                      1.000000          -0.002549      NaN   
DaysInHospital_Y2            -0.002549           1.000000      NaN   
predict                            NaN                NaN      NaN   
claims_count                  0.001116           0.207071      NaN   
providers_count               0.000478           0.199815      NaN   
PCPs_count                   -0.002911           0.038937      NaN   
vendors_count                 0.000747           0.194589      NaN   
placesvc_count                0.001009           0.166937      NaN   
specialities_count            0.000990           0.150488      NaN   
primaryconditionGroups_count  0.000296           0.188758      NaN   

                              claims_count  providers_count  PCPs_count  \
memberid                          0.001116         0.000478   -0.002911   
DaysInHospital_Y2                 0.207071         0.199815    0.038937   
predict                                NaN              NaN         NaN   
claims_count                      1.000000         0.888641    0.155317   
providers_count                   0.888641         1.000000    0.171819   
PCPs_count                        0.155317         0.171819    1.000000   
vendors_count                     0.842983         0.945981    0.172459   
placesvc_count                    0.711457         0.793960    0.153260   
specialities_count                0.806028         0.890804    0.148227   
primaryconditionGroups_count      0.850573         0.793414    0.134534   

                              vendors_count  placesvc_count  \
memberid                           0.000747        0.001009   
DaysInHospital_Y2                  0.194589        0.166937   
predict                                 NaN             NaN   
claims_count                       0.842983        0.711457   
providers_count                    0.945981        0.793960   
PCPs_count                         0.172459        0.153260   
vendors_count                      1.000000        0.804744   
placesvc_count                     0.804744        1.000000   
specialities_count                 0.886129        0.780742   
primaryconditionGroups_count       0.754058        0.656851   

                              specialities_count  primaryconditionGroups_count  
memberid                                0.000990                      0.000296  
DaysInHospital_Y2                       0.150488                      0.188758  
predict                                      NaN                           NaN  
claims_count                            0.806028                      0.850573  
providers_count                         0.890804                      0.793414  
PCPs_count                              0.148227                      0.134534  
vendors_count                           0.886129                      0.754058  
placesvc_count                          0.780742                      0.656851  
specialities_count                      1.000000                      0.751032  
primaryconditionGroups_count            0.751032                      1.000000  

[10 rows x 10 columns]

"""
def MemberSexDig():
    """
    Member
    Sex 
    F:42461 
    M:34828
    claims
    F:378963
    M:265743
    In [50]: m[m.sex=='F']['specialty'].value_counts()
    Out[50]: 
    Internal                     98531
    General Practice             74932
    Laboratory                   73700
    Diagnostic Imaging           42153
    Surgery                      27859
    Emergency                    18435
    Other                        12268
    Obstetrics and Gynecology    10393
    Pediatrics                    9188
    Rehabilitation                4542
    Anesthesiology                4472
    Pathology                     2490
    
    In [51]: m[m.sex=='M']['specialty'].value_counts()
    Out[51]: 
    Internal                     72111
    General Practice             54352
    Laboratory                   50625
    Surgery                      25915
    Diagnostic Imaging           24488
    Emergency                    13553
    Other                         8715
    Pediatrics                    8487
    Anesthesiology                3027
    Rehabilitation                2740
    Pathology                     1704
    Obstetrics and Gynecology       26

                  M      F
    ARTHSPIN  27324  44387
    CANCRM       62    169
    CATAST      205    240
    CHF        1584   1588
    COPD       5298   6408
    FLaELEC     422    825
    FXDISLC    4077   4605
    GIBLEED   10986  17860
    GIOBSENT   1239   1592
    GYNEC1      416  10818
    GYNECA       20   2349
    HEART4     3391   3686
    HEMTOL     2250   3944
    HIPFX       243    780
    INFEC4     9378  12163
    LIVERDZ     358    392
    METAB1      490    534
    METAB3    32166  39634
    MISCHRT   13293  18128
    MISCL1      515    786
    MISCL5     4864   7451
    MSC2a3    41470  70029
    NEUMENT   18719  27586
    ODaBNCA    4418   7898
    PERINTL      63     98
    PNCRDZ       98    161
    PNEUM      1268   1313
    PRGNCY      184   7649
    RESPR4    16149  21384
    ROAMI      6029   6806
    SEIZURE    2078   3084
    SEPSIS       55     65
    SKNAUT    11195  15706
    STROKE      996   1153
    TRAUMA     8994   9966
    UTI        2286   7869
    
    [36 rows x 2 columns]


    """
    m=Merge()
    F=m[m.sex=='F']['PrimaryConditionGroup'].value_counts()
    M=m[m.sex=='M']['PrimaryConditionGroup'].value_counts()
    #M.cumsum().plot()    
    MF=pd.concat([M,F],axis=1)
    MF[MF.M>MF.F]
    MF[MF.M<MF.F]