《数据科学实战：Python篇》八大直播作，第一次作业（附代码）

使用auto_ins作如下分析

1、首先对loss重新编码为1/0，有数值为1，命名为loss_flag

2、对loss_flag分布情况进行描述分析

3、分析是否出险和年龄、驾龄、性别、婚姻状态等变量之间的关系

# -*- coding: utf-8 -*-

"""

Created on Thu May 31 17:34:48 2018



@author: Young

"""



# In[1]:

import pandas as pd

import numpy as np

import matplotlib.pyplot as plt

import matplotlib

import seaborn as sns

import os



# In[2]:

os.chdir(r'D:\temp\homework')

df = pd.read_csv("auto_ins.csv",encoding="GBK")



# In[3]:

#增加loss_flag列

df["loss_flag"] = df["Loss"].apply(lambda x:0 if x==0 else 1)

df.head()



# In[4]

# 用饼图对loss_flag分布情况描述分析,autopct带百分数参数

df.loss_flag.value_counts().plot(kind="pie",autopct='%.2f%%')



# In[5]

# 出险和年龄的关系

sub_age_df = pd.crosstab(df.Age,df.loss_flag)

sub_age_df["sum1"]=sub_age_df.sum(1)

sub_age_df = sub_age_df.div(sub_age_df.sum1,axis=0)

sub_age_df[[0,1]].plot(kind = 'bar',stacked= True)



# In[5]

# 出险和驾龄的关系

sub_vage_df = pd.crosstab(df.vAge,df.loss_flag)

sub_vage_df["sum1"]=sub_vage_df.sum(1)

sub_vage_df = sub_vage_df.div(sub_vage_df.sum1,axis=0)

sub_vage_df.head()

sub_vage_df[[0,1]].plot(kind = 'bar',stacked= True)



# In[6]

# 出险和性别的关系

sub_gender_df = pd.crosstab(df.Gender,df.loss_flag)

sub_gender_df["sum1"]=sub_gender_df.sum(1)

sub_gender_df = sub_gender_df.div(sub_gender_df.sum1,axis=0)

sub_gender_df.head()

sub_gender_df[[0,1]].plot(kind = 'bar',stacked= True)



# In[7]

# 出险和婚否的关系

sub_marital_df = pd.crosstab(df.Marital,df.loss_flag)

sub_marital_df["sum1"]=sub_marital_df.sum(1)

sub_marital_df = sub_marital_df.div(sub_marital_df.sum1,axis=0)

sub_marital_df.head()

sub_marital_df[[0,1]].plot(kind = 'bar',stacked= True)