1 class Titanic(object):
  2     def __init__(self):
  3         self.data = titanic
  4 
  5     # 1.存活率是多少
  6     def rate_survive(self):
  7         survived = self.data.loc[:, 'Survived'].value_counts()[1]
  8         death = self.data.loc[:, 'Survived'].value_counts()[0]
  9         rate = float(survived) / (float(death) + float(survived))
 
 10         print '總人數：{},存活人數：{},死亡人數：{}'.format(survived + death, survived, death)
 11         return u'存活率：' + '%.2f' % rate
 12 
 13     # 2.哪個年齡段存活率最高
 14     def max_survive(self):
 15         age18_survived = self.data[self.data[u'Age'] <= 18][u'Survived'].value_counts()[1]
 16         age18_death = self.data[self.data[u'
 
Age'] <= 18][u'Survived'].value_counts()[0]
 17         age18_rate = float(age18_survived) / (float(age18_survived) + float(age18_death))
 18 
 19         age1860_survived = self.data[(self.data[u'Age'] > 18) & (self.data[u'Age'] < 60)][u'Survived'].value_counts()[1]
 20         age1860_death = self.data[(self.data[u'
 
Age'] > 18) & (self.data[u'Age'] < 60)][u'Survived'].value_counts()[0]
 21         age1860_rate = float(age1860_survived) / (float(age1860_survived) + float(age1860_death))
 22 
 23         age60_survived = self.data[self.data[u'Age'] >= 60][u'Survived'].value_counts()[1]
 24         age60_death = self.data[self.data[u'Age'] >= 60][u'Survived'].value_counts()[0]
 25         age60_rate = float(age60_survived) / (float(age60_survived) + float(age60_death))
 26 
 27         rate = [age18_rate, age60_rate, age1860_rate]
 28         age_data = ['18歲以下', '18-60歲', '60歲以上']
 29         max_rate = max(rate)
 30         age_range = age_data[rate.index(max(rate))]
 31         return '存活率最高的年齡段是{}，存活率為{}'.format(age_range, max_rate)
 32 
 33     # 3.女性存活率是否高於男性
 34     def than_survive(self):
 35         male_survied = self.data[self.data[u'Sex'] == u'male'][u'Survived'].value_counts()[1]
 36         male_death = self.data[self.data[u'Sex'] == u'male'][u'Survived'].value_counts()[0]
 37         rate_male = float(male_survied) / (float(male_survied) + float(male_death))
 38         print '男性共有{}人，存活{}人，死亡{}人'.format(male_death + male_survied, male_survied, male_death)
 39         female_survied = self.data[self.data[u'Sex'] == u'female'][u'Survived'].value_counts()[1]
 40         female_death = self.data[self.data[u'Sex'] == u'female'][u'Survived'].value_counts()[0]
 41         rate_female = float(female_survied) / (float(female_survied) + float(female_death))
 42         print '女性共有{}人，存活{}人，死亡{}人'.format(female_death + female_survied, female_survied, female_death)
 43         if rate_male > rate_female:
 44             return u'男性存活率更高，存活率為：%.2f' % rate_male
 45         else:
 46             return u'女性存活率更高，存活率為：%.2f' % rate_female
 47 
 48     # 4.船上是否出現貧富差距
 49     def poor_wealth(self):
 50         max_wealth = self.data[u'Fare'].max()
 51         max_poor = self.data[u'Fare'].min()
 52         if max_wealth - max_poor > 500:
 53             return '船上乘客最多消費了{}，最少消費了{}，存在貧富差距'.format(max_wealth, max_poor)
 54         else:
 55             return '船上乘客最多消費了{}，最少消費了{}，不存在貧富差距'.format(max_wealth, max_poor)
 56 
 57     # 5.頭等艙乘客的存活率是否高於經濟艙
 58     def pclass_survive(self):
 59         pclass1_survived = self.data[self.data[u'Pclass'] == 1]['Survived'].value_counts()[1]
 60         pclass1_death = self.data[self.data[u'Pclass'] == 1]['Survived'].value_counts()[0]
 61         pclass1_rate = float(pclass1_survived) / (float(pclass1_survived) + float(pclass1_death))
 62 
 63         pclass3_survived = self.data[self.data[u'Pclass'] == 3]['Survived'].value_counts()[1]
 64         pclass3_death = self.data[self.data[u'Pclass'] == 3]['Survived'].value_counts()[0]
 65         pclass3_rate = float(pclass3_survived) / (float(pclass3_survived) + float(pclass3_death))
 66 
 67         if pclass3_rate > pclass1_rate:
 68             return '頭等艙乘客存活率更高，存活率為{}'.format(pclass3_rate)
 69         else:
 70             return '經濟艙乘客存活率更高，存活率為{}'.format(pclass1_rate)
 71 
 72     # 6.有親屬在船上乘客比率，有親屬是否會影響存活率
 73     def family_survive(self):
 74         has_family = self.data[(self.data[u'Parch'] != 0) | (self.data[u'SibSp'] != 0)][u'PassengerId'].count()
 75         no_family = self.data[(self.data[u'Parch'] == 0) & (self.data[u'SibSp'] == 0)][u'PassengerId'].count()
 76         rate_family = float(has_family) / (float(has_family) + float(no_family))
 77 
 78         has_family_survived = \
 79             self.data[(self.data[u'Parch'] != 0) | (self.data[u'SibSp'] != 0)][u'Survived'].value_counts()[1]
 80         has_family_death = \
 81             self.data[(self.data[u'Parch'] != 0) | (self.data[u'SibSp'] != 0)][u'Survived'].value_counts()[0]
 82         has_family_rate = float(has_family_survived) / (float(has_family_survived) + float(has_family_death))
 83 
 84         no_family_survived = \
 85             self.data[(self.data[u'Parch'] == 0) & (self.data[u'SibSp'] == 0)][u'Survived'].value_counts()[1]
 86         no_family_death = \
 87             self.data[(self.data[u'Parch'] == 0) & (self.data[u'SibSp'] == 0)][u'Survived'].value_counts()[0]
 88         no_family_rate = float(no_family_survived) / (float(no_family_survived) + float(no_family_death))
 89 
 90         print '船上乘客中有親屬也在船上的有{}人，無親屬在船上的有{}人，有親屬在船上的乘客的比率為{}'.format(has_family, no_family, rate_family)
 91         if has_family_rate > no_family_rate:
 92             return '有親屬在船上的乘客存活率更高，存活率為{}'.format(has_family_rate)
 93         else:
 94             return '無親屬在船上的乘客存活率更高，存活率為{}'.format(no_family_rate)
 95 
 96     # 7.從哪個港口登船是否影響獲救
 97     def emarked_survive(self):
 98         Embarked_S_survived = self.data[self.data[u'Embarked'] == 'S'][u'Survived'].value_counts()[1]
 99         Embarked_S_death = self.data[self.data[u'Embarked'] == 'S'][u'Survived'].value_counts()[0]
100         Embarked_S_rate = float(Embarked_S_survived) / (float(Embarked_S_survived) + float(Embarked_S_death))
101 
102         Embarked_C_survived = self.data[self.data[u'Embarked'] == 'C'][u'Survived'].value_counts()[1]
103         Embarked_C_death = self.data[self.data[u'Embarked'] == 'C'][u'Survived'].value_counts()[0]
104         Embarked_C_rate = float(Embarked_C_survived) / (float(Embarked_C_survived) + float(Embarked_C_death))
105 
106         Embarked_Q_survived = self.data[self.data[u'Embarked'] == 'Q'][u'Survived'].value_counts()[1]
107         Embarked_Q_death = self.data[self.data[u'Embarked'] == 'Q'][u'Survived'].value_counts()[0]
108         Embarked_Q_rate = float(Embarked_Q_survived) / (float(Embarked_Q_survived) + float(Embarked_Q_death))
109 
110         embarked = ['S港口', 'C港口', 'Q港口']
111         rate = [Embarked_S_rate, Embarked_C_rate, Embarked_Q_rate]
112         max_rate = max(rate)
113         return '{}存活率最大，為{}'.format(embarked[rate.index(max_rate)], max_rate)
114 
115     # 8.不同年齡段女性的獲救率
116     def female_survive(self):
117         female18_survived = \
118             self.data[(self.data[u'Age'] <= 18) & (self.data[u'Sex'] == u'female')][u'Survived'].value_counts()[1]
119         female18_death = \
120             self.data[(self.data[u'Age'] <= 18) & (self.data[u'Sex'] == u'female')][u'Survived'].value_counts()[0]
121         female18_rate = float(female18_survived) / (float(female18_survived) + float(female18_death))
122 
123         female1850_survived = \
124             self.data[(self.data[u'Age'] > 18) & (self.data[u'Age'] < 50) & (self.data[u'Sex'] == u'female')][
125                 u'Survived'].value_counts()[1]
126         female1850_death = \
127             self.data[(self.data[u'Age'] > 18) & (self.data[u'Age'] < 50) & (self.data[u'Sex'] == u'female')][
128                 u'Survived'].value_counts()[0]
129         female1850_rate = float(female1850_survived) / (float(female1850_survived) + float(female1850_death))
130 
131         female50_survived = \
132             self.data[(self.data[u'Age'] >= 50) & (self.data[u'Sex'] == u'female')][u'Survived'].value_counts()[1]
133         female50_death = \
134             self.data[(self.data[u'Age'] >= 50) & (self.data[u'Sex'] == u'female')][u'Survived'].value_counts()[0]
135         female50_rate = float(female50_survived) / (float(female50_survived) + float(female50_death))
136         
137         return '18歲以下女性存活率：{}，18-50歲女性存活率：{}，50歲以上女性存活率：{}'.format(female18_rate, female1850_rate, female50_rate)
138 
139 
140 if __name__ == '__main__':
141     tt = Titanic()
142     # print tt.rate_survive()
143     # print tt.than_survive()
144     # print tt.max_survive()
145     # print tt.poor_wealth()
146     # print tt.pclass_survive()
147     # print tt.family_survive()
148     # print tt.emarked_survive()
149     print tt.female_survive()