python爬取空气质量数据，并用Matplotlib模块画出图

• 爬取到的数据

• 分析图：

• 源代码：

1.  
   import csv
2.  
   import random
3.  
   import time
4.  
   import pandas as pd
5.  
   import requests
6.  
   from bs4 import BeautifulSoup
7.  
   import matplotlib.pyplot as plt
8.  
    
9.  
   plt.rcParams["font.sans-serif"] = ["SimHei"]
10.  
    plt.rcParams["axes.unicode_minus"] = False
11.  
     
12.  
     
13.  
    class Spider(object):
14.  
    def __init__(self):
15.  
    pass
16.  
     
17.  
    def get_data_and_save(self):
18.  
    with open('data.csv', 'w', encoding='utf-8', newline='') as f:
19.  
    csv_writer = csv.writer(f)
20.  
    csv_writer.writerow(['日期', '质量等级', 'AQI指数', '当天AQI排名', 'PM2.5', 'PM10', 'So2', 'No2', 'Co', 'O3'])
21.  
    years = [2020, 2021, 2022]
22.  
    months = range(1, 13)
23.  
    for year in years:
24.  
    for month in months:
25.  
    if month < 10:
26.  
    url = f'http://tianqihoubao.com/aqi/beijing-{year}0{month}.html'
27.  
    else:
28.  
    url = f'http://tianqihoubao.com/aqi/beijing-{year}{month}.html'
29.  
    res = requests.get(url).text
30.  
    soup = BeautifulSoup(res, 'html.parser')
31.  
    for attr in soup.find_all('tr')[1:]:
32.  
    one_day_data = list()
33.  
    for index in range(0, 10):
34.  
    one_day_data.append(attr.find_all('td')[index].get_text().strip())
35.  
    csv_writer.writerow(one_day_data)
36.  
    time.sleep(2 random.random())
37.  
    print(year, month)
38.  
     
39.  
    def drawing(self):
40.  
    csv_df = pd.read_csv('data.csv', encoding='GBK')
41.  
    csv_df['日期'] = pd.to_datetime(csv_df['日期'])
42.  
    csv_df.index = csv_df['日期']
43.  
    del csv_df['日期']
44.  
     
45.  
    str_data = '2020-01-01'
46.  
    end_data = '2020-01-31'
47.  
    new_split_df = csv_df[str_data:end_data]
48.  
    values_count = new_split_df['质量等级'].value_counts().items()
49.  
    total = new_split_df['质量等级'].value_counts().tolist()
50.  
    label = new_split_df['质量等级'].value_counts().index
51.  
    explode = [0.01 for i in range(len(total))]
52.  
    fig, axes = plt.subplots(3, 1, figsize=(10, 8))
53.  
    # 画饼图
54.  
    axes[0].pie(total, labels=label, autopct='%0.2f%%', explode=explode)
55.  
    axes[0].set_title(f'北京{str_data.split("-")[0]}年{str_data.split("-")[1]}月份统计数据')
56.  
    # 画折线图
57.  
    axes[1].plot(new_split_df['AQI指数'], 'go-')
58.  
    axes[1].axhspan(0, 50, facecolor='g', alpha=0.5)
59.  
    axes[1].axhspan(50, 100, facecolor='y', alpha=0.5)
60.  
    axes[1].axhspan(100, 150, facecolor='orange', alpha=0.5)
61.  
    axes[1].axhspan(150, 200, facecolor='r', alpha=0.5)
62.  
    axes[1].set_title(f'北京{str_data.split("-")[0]}年{str_data.split("-")[1]}月空气质量指数（AQI）变化趋势')
63.  
    axes[1].legend(labels=('AQI',), loc='lower center', bbox_to_anchor=(0.5, -0.32))
64.  
    # 画折线图
65.  
    key = lambda x: f"{x.year}-{x.month}" if x.month >= 10 else f"{x.year}-0{x.month}"
66.  
    AQI_max = csv_df.groupby(key)['AQI指数'].max()
67.  
    AQI_ave = csv_df.groupby(key)['AQI指数'].mean()
68.  
    AQI_min = csv_df.groupby(key)['AQI指数'].min()
69.  
     
70.  
    axes[2].plot(AQI_max, 'go-', color='r')
71.  
    axes[2].plot(AQI_ave, 'go-', color='g')
72.  
    axes[2].plot(AQI_min, 'go-', color='b')
73.  
    axes[2].axhspan(0, 50, facecolor='g', alpha=0.5)
74.  
    axes[2].axhspan(50, 100, facecolor='y', alpha=0.5)
75.  
    axes[2].axhspan(100, 150, facecolor='orange', alpha=0.5)
76.  
    axes[2].axhspan(150, 200, facecolor='r', alpha=0.5)
77.  
    axes[2].axhspan(200, 300, facecolor='magenta', alpha=0.5)
78.  
    axes[2].axhspan(300, 400, facecolor='purple', alpha=0.5)
79.  
    axes[2].axhspan(400, 600, facecolor='brown', alpha=0.5)
80.  
    axes[2].legend(['最大值', '平均值', '最小值'], bbox_to_anchor=(0.6, -0.353), ncol=3)
81.  
    plt.xticks(rotation=70)
82.  
    plt.subplots_adjust(hspace=0.433, wspace=0.5)
83.  
    axes[2].set_title(f'北京空气质量指数（AQI）月变化趋势')
84.  
    plt.show()
85.  
     
86.  
     
87.  
    if __name__ == '__main__':
88.  
    spider = Spider()
89.  
    # 获取数据，保存数据
90.  
    spider.get_data_and_save()
91.  
    # 画分析图
92.  
    spider.drawing()

本案例只做教学用途，不可用于商业用途

这篇好文章是转载于：学新通技术网