image.png

两个数据集

将两个数据集联合起来

数据集的联合

代码

import os
import matplotlib.pyplot as plt
import pandas as pd

user_device_file = '/Users/miraco/PycharmProjects/DataMining/data_pd/mobile_data/user_device.csv'
user_usage_file  = '/Users/miraco/PycharmProjects/DataMining/data_pd/mobile_data/user_usage.csv'

output_path = '/Users/miraco/PycharmProjects/DataMining/output3'

#数据获取
user_device_df = pd.read_csv(user_device_file)
user_usage_df = pd.read_csv(user_usage_file)

#字符串列合并，把platform,platform_version两列合并，但是有一列是数字，要转换成字符
user_device_df['platform_version'] = user_device_df['platform_version'].astype('str')
user_device_df['system'] = user_device_df['platform'].str.cat(user_device_df['platform_version'],sep = '_')

#合并数据集
merged_df = pd.merge(user_device_df, user_usage_df, how = 'inner', on = 'user_id')

#数据分析
#按照系统查看每月流量的使用情况
system_usage_ser = merged_df.groupby('system')['monthly_mb'].mean()
#降序排列，自身接收自身的结果（不想创建新变量）
system_usage_ser.sort_values(ascending=False, inplace = True)

system_usage_ser.to_csv(os.path.join(output_path,'mobile_system_usage.csv'))
system_usage_ser.plot(kind = 'bar', rot = 45)
plt.ylabel('Monthly User (MB)')
plt.tight_layout()
plt.savefig(os.path.join(output_path,'mobile_system_usage.png'))
plt.show()

运行出图

降序的按照系统的流量使用降序排列图

练习：统计不同专业的员工平均薪资

• 题目描述：统计不同专业背景的员工的平均薪资，并用柱状图显示结果

• 题目要求:

• 使用Pandas进行数据分析及可视化

• 数据文件：

• 数据源下载地址：https://video.mugglecode.com/data_employee.zip，下载压缩包后解压即可

• employee_info.csv，包含了员工的基本信息，共4列数据，分别表示：

1. EmployeeNumber: 员工编号
2. Age: 年龄
3. Department: 所处部门
4. MonthlyIncome: 月收入

• employee_edu.csv，包含部分员工的专业背景，共2列数据：
• EmployeeNumber: 员工编号
• EducationField: 专业背景

import os
import matplotlib.pyplot as plt
import pandas as pd

employee_edu_file = '/Users/miraco/PycharmProjects/DataMining/data_pd/data_employee/employee_edu.csv'
employee_info_file  = '/Users/miraco/PycharmProjects/DataMining/data_pd/data_employee/employee_info.csv'

output_path = '/Users/miraco/PycharmProjects/DataMining/output3'

#数据获取
employee_edu_df = pd.read_csv(employee_edu_file)
employee_info_df = pd.read_csv(employee_info_file)

#数据处理
merged_df = pd.merge(employee_edu_df, employee_info_df, how = 'inner',on = 'EmployeeNumber')

#数据分析
seq = merged_df.groupby('EducationField')['MonthlyIncome'].mean().sort_values()

#数据展示
seq.plot(kind = 'bar', rot = 45)
plt.ylabel('Monthly Income')
plt.xlabel('Education Field')
plt.tight_layout()
plt.show()

运行结果

image.png