基于爬取百合网的数据，用matplotlib生成图表

爬取百合网的数据链接：http://www.cnblogs.com/YuWeiXiF/p/8439552.html

总共爬了22779条数据。第一次接触matplotlib库，以下代码参考了matplotlib官方文档：https://matplotlib.org/users/index.html。

数据查询用到了两个方法：getSexNumber(@sex varchar(2),@income varchar(30))、gethousingNumber(@sex varchar(2),@housing varchar(6))来简化查询语句的长度，代码如下：

 go
 create function getSexNumber(@sex varchar(),@income varchar())
 returns int
 as
 begin
     return(select count(id) from users where sex = @sex and income = @income)
 end
 go
 go
 create function gethousingNumber(@sex varchar(),@housing varchar())
 returns int
 as
 begin
     return(select count(id) from users where sex = @sex and housing = @housing)
 end
 go

 以下代码为SQL Server 数据库操作：

 #__author: "YuWei"
 #__date: 2018/2/11
 import numpy as np
 import matplotlib.pyplot as plt
 import pymssql

 def db(sql):
     """
     数据库相关操作

     :param sql: sql语句
     :return: 查询的结果集，list封装
     """
     conn = pymssql.connect(host='localhost', user='sa', password='123456c', database='Baihe', charset="utf8")
     cur = conn.cursor()
     cur.execute(sql)
     row = cur.fetchone() # 指向结果集的第一行，
     data = [] # 返回的list
     while row:
         rows = list(row)
         for i in range(len(rows)): # 针对rows的每项编码
             try:
                 rows[i] = rows[i].encode('latin-1').decode('gbk')
             except AttributeError:pass
         data.append(rows) # 向data加数据
         row = cur.fetchone() #
     print(data)
     cur.close()
     conn.close()
     return data

生成各工资段人数占总人数比图:

 def builder_income_ratio():
     """
     生成各工资段人数占总人数比图

     :return: 无
     """
     data_list = db("select income,count(id) from users group by income")
     income_data_list = [] # 数据
     income_labels_list = [] # 图例
     for data in data_list:
         income_data_list.append(data[1])
         income_labels_list.append(data[0])
     income_data_list.remove(income_data_list[6]) # 删掉不要的数据
     income_labels_list.remove(income_labels_list[6]) # 删掉不要的数据
     # 画饼图
     plt.pie(income_data_list,labels=income_labels_list,colors=['c','m','r','g'],startangle=30,
             shadow=True,explode=(0, 0, 0.1, 0, 0, 0, 0.1, 0, 0.1, 0, 0, 0),autopct='%.1f%%')
     plt.title('各工资段人数占总人数比') # 标题
     plt.show() # 显示

执行效果如下：

生成各工资段男，女人数图：

 def builder_sex_ratio():
     """
     生成各工资段男，女人数图

     :return: 无
     """
     data_list = db("select income,dbo.getSexNumber('男',income) as 男 ,dbo.getSexNumber('女',income) as 女 "
                     "from users group by income")
     men = [] # 男
     women = [] # 女
     labels =[] # 图例
     for data in data_list:
         labels.append(data[0])
         men.append(data[1])
         women.append(data[2])
     men.remove(men[6]) # 删掉不要的数据
     women.remove(women[6]) # 删掉不要的数据
     labels.remove(labels[6]) # 删掉不要的数据
     max_line = 12 # 12个
     fig,ax = plt.subplots()
     line = np.arange(max_line) # [0,1,2,3,4,5,6,7,8,9,10,11]
     bar_width = 0.4 # 条形之间的宽度
     # 画条形图
     ax.bar(line, men, bar_width,alpha=0.3, color='b',label='男')
     ax.bar(line+bar_width, women, bar_width,alpha=0.3, color='r',label='女')
     ax.set_xlabel('工资段')
     ax.set_ylabel('人数')
     ax.set_title('各工资段男，女人数图')
     ax.set_xticks(line + bar_width / 2) # 保证条形居中
     ax.set_xticklabels(labels)
     # 画两条线
     plt.plot([0.04, 1.04, 2.04, 3.04, 4.04, 5.04, 6.04, 7.04, 8.04, 9.04, 10.04, 11.04], men, label='男')
     plt.plot([0.4, 1.4, 2.4, 3.4, 4.4, 5.4, 6.4, 7.4, 8.4, 9.4, 10.4, 11.4], women, label='女')
     ax.legend()
     fig.tight_layout()
     # fig.savefig("1.png") # 生成图片
     plt.show()

执行效果如下：

生成男，女平均身高图：

 def builder_age_ratio():
     """
     生成男，女平均身高图

     :return:
     """
     data_list = db("select sex,avg(height) as 平均升高 from users group by sex")
     sex = [] # 性别
     number = [] # 人数
     for data in data_list:
         sex.append(data[0])
         number.append(data[1])
     # 画条形图
     plt.bar(sex[0], number[0], label="男", color='g',width=0.03)
     plt.bar(sex[1], number[1], label="女", color='r',width=0.03)
     plt.legend()
     plt.xlabel('性别')
     plt.ylabel('身高')
     plt.title('男女平均身高图')
     plt.show()

执行效果如下：

生成有房与无房的人数比例图：

 def builder_housing_sum_ratio():
     """
     生成有房与无房的人数比例图

     :return:
     """
     data_list = db("select housing,count(id) from users group by housing")
     housing_data_list = []
     housing_labels_list = []
     for data in data_list:
         housing_data_list.append(data[1])
         housing_labels_list.append(data[0])
     # 画饼图
     plt.pie(housing_data_list, labels=housing_labels_list, colors=['g', 'r'], startangle=30,
             shadow=True, explode=(0, 0), autopct='%.0f%%')
     plt.title('有房与无房的人数比例图')
     plt.show()

执行效果如下：

生成有无房男女人数图：

 def builder_housing_ratio():
     """
     生成有无房男女人数图

     :return:
     """
     data_list = db("select dbo.gethousing('女',housing),dbo.gethousing('男',housing) from users group by housing")
     homey = [] # 有房
     homem = [] # 无房
     for data in data_list:
         homey.append(data[0])
         homem.append(data[1])
     max_line = 2 # 两个
     fig, ax = plt.subplots()
     line = np.arange(max_line) # [0,1]
     bar_width = 0.1 # 条形之间的宽度
     # 画条形
     ax.bar(line,homey , bar_width, alpha=0.3,color='b',label='女')
     ax.bar(line+bar_width, homem, bar_width,alpha=0.3,color='r',label='男')
     ax.set_xlabel('有无房')
     ax.set_ylabel('人数')
     ax.set_title('有无房男女人数图')
     ax.set_xticks(line + bar_width / 2) # 保持居中
     ax.set_xticklabels(['有房','无房'])
     ax.legend()
     fig.tight_layout()
     plt.show()

执行效果如下：