分析目的：

通过分析训练数据集分析出什么类型的人更可能生存，并预测出数据集中乘客的生还概率

#字段说明

# PassengerId： 乘客编号
# Survived：是否生存，也是我们分析目标。1表示生存，0表示死亡
# Pclass：船舱等级 分1、2、3等级，1等级最高
# Name：乘客姓名
# Sex：性别
# Age：年龄
# SibSp：该乘客一起旅行的兄弟姐妹和配偶的数量(同代直系亲属人数)
# Parch：该和乘客一起旅行的父母和孩子的数量(不同代直系亲属人数)
# Ticket：船票号
# Fare：船票价格
# Cabin：船舱号
# Embarked：登船港口
# S=英国南安普顿Southampton(起航点)
# C=法国 瑟堡市Cherbourg(途经点)
# Q=爱尔兰 昆士敦Queenstown(途经点)

1.获取数据

import pandas as pd 
import numpy as np
import matplotlib.pyplot as plt
#获取数据
titanic_train=pd.read_csv("./titanic_train.csv")
titanic_test=pd.read_csv("./titanic_test.csv")

2.数据探索

titanic_train.head()#前5个样本

titanic_test.head()

titanic_train.info()
titanic_test.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 891 entries, 0 to 890
Data columns (total 12 columns):
 #   Column       Non-Null Count  Dtype  
---  ------       --------------  -----  
 0   PassengerId  891 non-null    int64  
 1   Survived     891 non-null    int64  
 2   Pclass       891 non-null    int64  
 3   Name         891 non-null    object 
 4   Sex          891 non-null    object 
 5   Age          714 non-null    float64
 6   SibSp        891 non-null    int64  
 7   Parch        891 non-null    int64  
 8   Ticket       891 non-null    object 
 9   Fare         891 non-null    float64
 10  Cabin        204 non-null    object 
 11  Embarked     889 non-null    object 
dtypes: float64(2), int64(5), object(5)
memory usage: 83.7+ KB
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 418 entries, 0 to 417
Data columns (total 11 columns):
 #   Column       Non-Null Count  Dtype  
---  ------       --------------  -----  
 0   PassengerId  418 non-null    int64  
 1   Pclass       418 non-null    int64  
 2   Name         418 non-null    object 
 3   Sex          418 non-null    object 
 4   Age          332 non-null    float64
 5   SibSp        418 non-null    int64  
 6   Parch        418 non-null    int64  
 7   Ticket       418 non-null    object 
 8   Fare         417 non-null    float64
 9   Cabin        91 non-null     object 
 10  Embarked     418 non-null    object 
dtypes: float64(2), int64(4), object(5)
memory usage: 36.0+ KB

titanic_train.describe().T

titanic_train.describe(include=['O']).T

t_t_pclass=titanic_train[['Pclass', 'Survived']].groupby(['Pclass'], as_index=False)
t_t_pclass.mean().sort_values(by='Survived', ascending=False)

titanic_train[["Sex", "Survived"]].groupby(['Sex'], as_index=False).mean().sort_values(by='Survived', ascending=False)

import seaborn as sns
import matplotlib.pyplot as plt
age_map = sns.FacetGrid(titanic_train, col='Survived')
age_map.map(plt.hist, 'Age', bins=20)

<seaborn.axisgrid.FacetGrid at 0x1f9ebb158c8>

output_13_1.png

grid = sns.FacetGrid(titanic_train, col='Survived', row='Pclass', height=2.2, aspect=1.6)
grid.map(plt.hist, 'Age', alpha=.5, bins=20)
grid.add_legend()

<seaborn.axisgrid.FacetGrid at 0x1f9ee46c708>

output_14_1.png

3.特征工程

combine = [titanic_train, titanic_test]
for dataset in combine:
    dataset['Sex'] = dataset['Sex'].map( {'female': 1, 'male': 0} ).astype(int)
titanic_train.head()

for dataset in combine:    
    dataset['Age'].fillna(dataset['Age'].median(), inplace = True)
    dataset['Embarked'].fillna(dataset['Embarked'].mode()[0],    inplace = True)
    dataset['Fare'].fillna(dataset['Fare'].median(), inplace = True)
titanic_train.drop(['PassengerId','Cabin', 'Ticket'], axis=1, inplace = True)
titanic_train.info()
titanic_test.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 891 entries, 0 to 890
Data columns (total 9 columns):
 #   Column    Non-Null Count  Dtype  
---  ------    --------------  -----  
 0   Survived  891 non-null    int64  
 1   Pclass    891 non-null    int64  
 2   Name      891 non-null    object 
 3   Sex       891 non-null    int32  
 4   Age       891 non-null    float64
 5   SibSp     891 non-null    int64  
 6   Parch     891 non-null    int64  
 7   Fare      891 non-null    float64
 8   Embarked  891 non-null    object 
dtypes: float64(2), int32(1), int64(4), object(2)
memory usage: 59.3+ KB
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 418 entries, 0 to 417
Data columns (total 11 columns):
 #   Column       Non-Null Count  Dtype  
---  ------       --------------  -----  
 0   PassengerId  418 non-null    int64  
 1   Pclass       418 non-null    int64  
 2   Name         418 non-null    object 
 3   Sex          418 non-null    int32  
 4   Age          418 non-null    float64
 5   SibSp        418 non-null    int64  
 6   Parch        418 non-null    int64  
 7   Ticket       418 non-null    object 
 8   Fare         418 non-null    float64
 9   Cabin        91 non-null     object 
 10  Embarked     418 non-null    object 
dtypes: float64(2), int32(1), int64(4), object(4)
memory usage: 34.4+ KB

for dataset in combine:    
    dataset['FamilySize'] = dataset ['SibSp'] + dataset['Parch'] + 1
    dataset['IsAlone'] = 1 
    dataset['IsAlone'].loc[dataset['FamilySize'] > 1] = 0
    dataset['FareBin'] = pd.qcut(dataset['Fare'], 4)
    dataset['AgeBin'] = pd.cut(dataset['Age'].astype(int), 5)

C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\indexing.py:670: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  self._setitem_with_indexer(indexer, value)

from sklearn.preprocessing import OneHotEncoder, LabelEncoder
label = LabelEncoder()
for dataset in combine:    
    dataset['Sex_Code'] = label.fit_transform(dataset['Sex'])
    dataset['Embarked_Code'] = label.fit_transform(dataset['Embarked'])
    dataset['AgeBin_Code'] = label.fit_transform(dataset['AgeBin'])
    dataset['FareBin_Code'] = label.fit_transform(dataset['FareBin'])
titanic_train.head()

4.模型开发

Target = ['Survived']
titanic_train_x_bin = ['Sex','Pclass', 'Embarked_Code', 'FamilySize', 'AgeBin_Code', 'FareBin_Code']
titanic_train_xy_bin = Target + titanic_train_x_bin
print('Bin X Y: ', titanic_train_xy_bin, '\n')
#Model Algorithms
from sklearn import linear_model
#Common Model Helpers
from sklearn.preprocessing import OneHotEncoder, LabelEncoder
from sklearn import feature_selection
from sklearn import model_selection
from sklearn import metrics
#Visualization
import matplotlib as mpl
import matplotlib.pyplot as plt
import matplotlib.pylab as pylab
import seaborn as sns
train1_x_bin,test1_x_bin,train1_y_bin,test1_y_bin =model_selection.train_test_split(titanic_train[titanic_train_x_bin],titanic_train[Target],test_size=0.3,random_state = 0)
print("Data1 Shape: {}".format(titanic_train.shape))
print("Train1 Shape: {}".format(train1_x_bin.shape))
print("Test1 Shape: {}".format(test1_x_bin.shape))
train1_x_bin.head()

Bin X Y:  ['Survived', 'Sex', 'Pclass', 'Embarked_Code', 'FamilySize', 'AgeBin_Code', 'FareBin_Code'] 

Data1 Shape: (891, 17)
Train1 Shape: (623, 6)
Test1 Shape: (268, 6)

from sklearn import linear_model
logreg = linear_model.LogisticRegression()
logreg.fit(train1_x_bin, train1_y_bin)
Y_pred = logreg.predict(test1_x_bin)

C:\ProgramData\Anaconda3\lib\site-packages\sklearn\utils\validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().
  y = column_or_1d(y, warn=True)

print(logreg.intercept_) #截距
coef=pd.DataFrame(logreg.coef_).T  #参数
columns=pd.DataFrame(train1_x_bin.columns,columns=['A'])
result = pd.concat([columns,coef], axis=1)
result = result.rename(columns={'A': 'Attribute', 0: 'Coefficients'})
result

[1.82313942]

5.模型评估

acc_train = round(logreg.score(train1_x_bin, train1_y_bin) * 100, 2)
acc_train

78.81

# View summary of common classification metrics
print("------------------------Metrices---------------------")
print(metrics.classification_report(y_true = test1_y_bin,y_pred = Y_pred))

------------------------Metrices---------------------
              precision    recall  f1-score   support

           0       0.85      0.84      0.84       168
           1       0.74      0.75      0.74       100

    accuracy                           0.81       268
   macro avg       0.79      0.79      0.79       268
weighted avg       0.81      0.81      0.81       268

from sklearn.metrics import roc_curve, auc
import matplotlib.pyplot as plt
probs = logreg.predict_proba(test1_x_bin)
preds = probs[:,1]
#---find the FPR, TPR, and threshold---
fpr, tpr, threshold = roc_curve(test1_y_bin, preds)
#---find the area under the curve---
roc_auc = auc(fpr, tpr)
plt.plot(fpr, tpr, 'b', label = 'AUC = %0.2f' % roc_auc)
plt.plot([0, 1], [0, 1],'r--')
plt.xlim([0, 1])
plt.ylim([0, 1])
plt.ylabel('True Positive Rate (TPR)')
plt.xlabel('False Positive Rate (FPR)')
plt.title('Receiver Operating Characteristic (ROC)')
plt.legend(loc = 'lower right')
plt.show()

output_27_0.png

通过分析得出以下结论：
性别方面：女性的生还率较高，达到74.2%；
年龄方面：4岁以下的婴儿存活率较高，15-25岁的人大部分没有生还。
船层方面：第一船层的乘客大部分生还，而第三船层的乘客大部分没有生还。

模型的准确率为78.49%，模型的AUC等于0.86，这个模型效果良好，能够很好预测
乘客的生存概率。