Python数据分析可视化--Titanic

谁借莪１个温暖的怀抱￠ 2023-01-07 09:00 307阅读 0赞

# Python数据分析可视化–Titanic #

> 这篇文章主要介绍泰坦尼克幸存者问题的数据处理以及可视乎部分，关于机器学习部分:  
> [机器学习2：KNN决策树探究泰坦尼克号幸存者问题][2_KNN]

### 文章目录 ###

*  Python数据分析可视化--Titanic
 *  导入数据
 *  数据探索
 *   *  判断是否存在缺失值
     *  关系探索
     *   *  仓位和存活率关系
         *  性别和存活率关系
         *  兄弟姐妹和孩子对于存活关系
         *  数据可视化分析
     *  数据预处理

import pandas as pd
    import seaborn as sns
    import matplotlib.pyplot as plt
    from matplotlib.pylab import style
    style.use('ggplot')
    import numpy as np
    from sklearn.linear_model import LogisticRegression
    from sklearn.svm import SVC,LinearSVC
    import warnings
    warnings.filterwarnings("ignore")

# 导入数据 #

df = pd.read_csv('train.csv',index_col=None)
    test_df = pd.read_csv('test.csv')

# 数据探索 #

## 判断是否存在缺失值 ##

np.any(pd.isnull(df))

True

np.any(df["Embarked"].isnull())

True

df.describe()

<table> 
 <thead> 
  <tr> 
   <th></th> 
   <th>PassengerId</th> 
   <th>Survived</th> 
   <th>Pclass</th> 
   <th>Age</th> 
   <th>SibSp</th> 
   <th>Parch</th> 
   <th>Fare</th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <th>count</th> 
   <td>891.000000</td> 
   <td>891.000000</td> 
   <td>891.000000</td> 
   <td>714.000000</td> 
   <td>891.000000</td> 
   <td>891.000000</td> 
   <td>891.000000</td> 
  </tr> 
  <tr> 
   <th>mean</th> 
   <td>446.000000</td> 
   <td>0.383838</td> 
   <td>2.308642</td> 
   <td>29.699118</td> 
   <td>0.523008</td> 
   <td>0.381594</td> 
   <td>32.204208</td> 
  </tr> 
  <tr> 
   <th>std</th> 
   <td>257.353842</td> 
   <td>0.486592</td> 
   <td>0.836071</td> 
   <td>14.526497</td> 
   <td>1.102743</td> 
   <td>0.806057</td> 
   <td>49.693429</td> 
  </tr> 
  <tr> 
   <th>min</th> 
   <td>1.000000</td> 
   <td>0.000000</td> 
   <td>1.000000</td> 
   <td>0.420000</td> 
   <td>0.000000</td> 
   <td>0.000000</td> 
   <td>0.000000</td> 
  </tr> 
  <tr> 
   <th>25%</th> 
   <td>223.500000</td> 
   <td>0.000000</td> 
   <td>2.000000</td> 
   <td>20.125000</td> 
   <td>0.000000</td> 
   <td>0.000000</td> 
   <td>7.910400</td> 
  </tr> 
  <tr> 
   <th>50%</th> 
   <td>446.000000</td> 
   <td>0.000000</td> 
   <td>3.000000</td> 
   <td>28.000000</td> 
   <td>0.000000</td> 
   <td>0.000000</td> 
   <td>14.454200</td> 
  </tr> 
  <tr> 
   <th>75%</th> 
   <td>668.500000</td> 
   <td>1.000000</td> 
   <td>3.000000</td> 
   <td>38.000000</td> 
   <td>1.000000</td> 
   <td>0.000000</td> 
   <td>31.000000</td> 
  </tr> 
  <tr> 
   <th>max</th> 
   <td>891.000000</td> 
   <td>1.000000</td> 
   <td>3.000000</td> 
   <td>80.000000</td> 
   <td>8.000000</td> 
   <td>6.000000</td> 
   <td>512.329200</td> 
  </tr> 
 </tbody> 
</table>

## 关系探索 ##

### 仓位和存活率关系 ###

df[['Pclass','Survived']].groupby(by=["Pclass"]).mean().sort_values(by="Survived",ascending=False)

<table> 
 <thead> 
  <tr> 
   <th></th> 
   <th>Survived</th> 
  </tr> 
  <tr> 
   <th>Pclass</th> 
   <th></th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <th>1</th> 
   <td>0.629630</td> 
  </tr> 
  <tr> 
   <th>2</th> 
   <td>0.472826</td> 
  </tr> 
  <tr> 
   <th>3</th> 
   <td>0.242363</td> 
  </tr> 
 </tbody> 
</table>

从表可知，仓位高低与存活率存在一定的关系

### 性别和存活率关系 ###

df['Sex'].value_counts()

male      577
    female    314
    Name: Sex, dtype: int64

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

<table> 
 <thead> 
  <tr> 
   <th></th> 
   <th>Survived</th> 
  </tr> 
  <tr> 
   <th>Sex</th> 
   <th></th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <th>female</th> 
   <td>0.742038</td> 
  </tr> 
  <tr> 
   <th>male</th> 
   <td>0.188908</td> 
  </tr> 
 </tbody> 
</table>

从表可知，女性的存活率明显高于男性，性别与存活率存在一定的关系

### 兄弟姐妹和孩子对于存活关系 ###

df[['SibSp','Survived']].groupby(by=["SibSp"]).mean().sort_index(ascending=False)

<table> 
 <thead> 
  <tr> 
   <th></th> 
   <th>Survived</th> 
  </tr> 
  <tr> 
   <th>SibSp</th> 
   <th></th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <th>8</th> 
   <td>0.000000</td> 
  </tr> 
  <tr> 
   <th>5</th> 
   <td>0.000000</td> 
  </tr> 
  <tr> 
   <th>4</th> 
   <td>0.166667</td> 
  </tr> 
  <tr> 
   <th>3</th> 
   <td>0.250000</td> 
  </tr> 
  <tr> 
   <th>2</th> 
   <td>0.464286</td> 
  </tr> 
  <tr> 
   <th>1</th> 
   <td>0.535885</td> 
  </tr> 
  <tr> 
   <th>0</th> 
   <td>0.345395</td> 
  </tr> 
 </tbody> 
</table>

df[['Parch','Survived']].groupby(by=["Parch"]).mean().sort_index(ascending=False)

<table> 
 <thead> 
  <tr> 
   <th></th> 
   <th>Survived</th> 
  </tr> 
  <tr> 
   <th>Parch</th> 
   <th></th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <th>6</th> 
   <td>0.000000</td> 
  </tr> 
  <tr> 
   <th>5</th> 
   <td>0.200000</td> 
  </tr> 
  <tr> 
   <th>4</th> 
   <td>0.000000</td> 
  </tr> 
  <tr> 
   <th>3</th> 
   <td>0.600000</td> 
  </tr> 
  <tr> 
   <th>2</th> 
   <td>0.500000</td> 
  </tr> 
  <tr> 
   <th>1</th> 
   <td>0.550847</td> 
  </tr> 
  <tr> 
   <th>0</th> 
   <td>0.343658</td> 
  </tr> 
 </tbody> 
</table>

兄弟姐妹和孩子对于存活关系不大

### 数据可视化分析 ###

1.  年龄和存活率关系

g1 = sns.FacetGrid(df,col="Survived")
    g1.map(plt.hist,"Age",bins=20)
    plt.show()

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ1MTc2NTQ4_size_16_color_FFFFFF_t_70_pic_center]

1.  船票等级和存活率关系

g2 = sns.FacetGrid(df,col="Survived",row="Pclass")
    g2.map(plt.hist,"Age")
    plt.show()

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ1MTc2NTQ4_size_16_color_FFFFFF_t_70_pic_center 1]

可以看到,船票等级和存活率存在关系

1.  复合关系

g3 = sns.FacetGrid(df,row="Embarked",height=3,aspect=1.6)
    g3.map(sns.pointplot,"Pclass",'Survived','Sex',palette="deep",hue_order=["female","male"])
    g3.add_legend()
    plt.show()

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ1MTc2NTQ4_size_16_color_FFFFFF_t_70_pic_center 2]

g4 = sns.FacetGrid(df,row="Embarked",col="Survived",height=3,aspect=1.6)
    g4.map(sns.barplot,'Sex','Fare',palette="deep",hue_order=["female","male"])
    g4.add_legend()
    plt.show()

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ1MTc2NTQ4_size_16_color_FFFFFF_t_70_pic_center 3]

*  票价和存活率存在关系，登船口和存活率也存在关系

## 数据预处理 ##

*  对于甲板层的分布对与于获救的分析，过于复杂，这里就不探究了

df = df.drop(["Ticket","Cabin"],axis=1)

*  缺失值填充

df['Age'].fillna(method="pad",inplace=True)
    df["Embarked"].fillna(method="pad",inplace=True)

*  分类数据转化

df["Embarked"]

0      S
    1      C
    2      S
    3      S
    4      S
          ..
    886    S
    887    S
    888    S
    889    C
    890    Q
    Name: Embarked, Length: 891, dtype: object

"""性别转换"""
    df["Sex"] = df["Sex"].map({
        "female":1,"male":0}).astype(int)
    #df.loc[df['Sex'] == 'male', 'Sex'] = 1       # 用数值1来代替male，用0来代替female
    #df.loc[df['Sex'] == 'female', 'Sex'] = 0
    """登船口转换"""
    df["Embarked"] = df["Embarked"].map({
        "S":0,"C":1,"Q":2}).astype(int)

[2_KNN]: https://blog.csdn.net/qq_45176548/article/details/112060492
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ1MTc2NTQ4_size_16_color_FFFFFF_t_70_pic_center]: /images/20221119/8cd849bcbbb54bcfb47a4447e5ed7765.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ1MTc2NTQ4_size_16_color_FFFFFF_t_70_pic_center 1]: /images/20221119/22d6db5051af4e89aa4ac529ea73486e.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ1MTc2NTQ4_size_16_color_FFFFFF_t_70_pic_center 2]: /images/20221119/333deb49ac2d43c6a046654ab4f6be61.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzQ1MTc2NTQ4_size_16_color_FFFFFF_t_70_pic_center 3]: /images/20221119/bc314e8e8419437a87b6dd726725f241.png