Jupyter Notebook + Pyecharts——学习笔记(04)

数据可视化案例

城市天气案例：

空气质量.xlsx
导出下载为本地文件
然后在Jupyter的home页面导入文件，就会与你的代码文件在同一目录下

import pandas as pd
import json
from pyecharts.charts import *
import pyecharts.options as opts
from pyecharts.globals import ThemeType, ChartType
1
2
3
4
5

导入数据

data = pd.read_excel("./空气质量.xlsx")
1

data.head()  #.head( ) ：是指取数据的前n行数据，默认是前5行（没有print语句）
1

数据去重

weather_level = data['质量等级'].unique() #.unique() 去重且不改变顺序
1

data['质量等级'].value_counts() #value_counts()查看表格某列中有多少个不同值，并计算每个值的重复次数。
1

quality_dict = {weather_level[i]:i+1 for i in range(len(weather_level))}# 用数字代表污染程度
1

quality_dict
1

data['level'] = data['质量等级'].map(quality_dict) #加了level一列
1

data.head()
1

city_list = data['城市'].unique()
province_list = data['省'].unique()
time_list = data['日期'].unique()
1
2
3

可检查数据

1、空气质量等级的地理分布：

国家地图展示不出来，就不放截图了

n =32

def timeline1():
    # 创建轮播图
    timeline = Timeline(init_opts = opts.InitOpts(width='1200px', height='700px'))
    for t in time_list[:n]:
        data_one_day = []
        # 取出所有城市当天空气污染质量值
        for i in city_list:
            index_c = (data['城市'] == i) & (data['日期'] == t)
            data_one_day.append((i, int(data.loc[index_c, 'level'])))

        geo = (Geo()
              .add_schema(maptype="china", itemstyle_opts=opts.ItemStyleOpts(border_color="#f7f8fa"))
              .add("", data_one_day, type_=ChartType.EFFECT_SCATTER)
              .set_series_opts(itemstyle_opts=opts.ItemStyleOpts(color='green'),
                              label_opts=opts.LabelOpts(is_show=False))
              .set_global_opts(title_opts=opts.TitleOpts(title="全国空气质量等级",
                                                        subtitle="{}更新".format(t),
                                                        pos_left='center'),
                              visualmap_opts=opts.VisualMapOpts(is_show=True, min_ = 0, max_ = 6, precision = 1,
                                                              is_piecewise=True, split_number=6,
                                                              pieces=[
                                                                  {"max":6, "min":5, "label":"严重污染",
                                                                  "color":"#B40404"},
                                                                  {"max":5, "min":4, "label":"重度污染",
                                                                  "color":"#DF0101"},
                                                                  {"max":4, "min":3, "label":"中度污染",
                                                                  "color":"#F78181"},
                                                                  {"max":3, "min":2, "label":"轻度污染",
                                                                  "color":"#F5A9A9"},
                                                                  {"max":2, "min":1, "label":"良",
                                                                  "color":"#b7dd6f"},
                                                                  {"max":1, "min":0, "label":"优",
                                                                  "color":"#32e0c4"},
                                                              ])
                              )
              )
        timeline.add(geo, t)
    timeline.add_schema(
        play_interval = 3000, #轮播速度
        is_timeline_show = True,#是否显示timeline组件
        is_auto_play=True
    )
    return timeline
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

timeline1().render_notebook()
1

2、空气质量最差的十个城市

def timeline2():
    timeline = Timeline(init_opts = opts.InitOpts(width = "600px", height = "600px"))
    for ind, t in enumerate(time_list[:n]):
        data_t  = data.loc[data['日期'] == t,:]
        #取出空气质量最差的十个城市
        data_top10 = data_t.sort_values(by='AQI指数', ascending = True).tail(10)
        #放入图
        
        bar = (Bar()
               .add_xaxis(data_top10['城市'].tolist())
               .add_yaxis('PM2.5', data_top10['PM2.5'].tolist(), stack='stack1')
               .add_yaxis('PM10', data_top10['PM10'].tolist(), stack='stack1')
               .add_yaxis('So2', data_top10['So2'].tolist(), stack='stack1')
               .add_yaxis('No2', data_top10['No2'].tolist(), stack='stack1')
               .reversal_axis()#转置
               .set_series_opts(label_opts=opts.LabelOpts(is_show=False))
               .set_global_opts(
                   title_opts = opts.TitleOpts(title = "空气质量最差的10个城市情况",
                                              pos_left='center'),
                   xaxis_opts=opts.AxisOpts(axislabel_opts=opts.LabelOpts(is_show=False),
                                            axisline_opts=opts.AxisLineOpts(is_show=False),
                                            axistick_opts=opts.AxisTickOpts(is_show=False)
                                           ),
                   legend_opts=opts.LegendOpts(pos_top="6%")
               )
              )
        timeline.add(bar, t)
    #配置轮播图
    timeline.add_schema(
        play_interval=3000, # 轮播速度
        is_timeline_show=True, #是否显示timeline组件
        is_auto_play=True
    )
    return timeline
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34

timeline().render_notebook()
1

3、8月份天气变化情况

def line():
    data_group = pd.pivot_table(data, 
                                index="日期", 
                                columns="质量等级",
                                values ="城市",
                                aggfunc=len,
                               fill_value=0)
    line = (Line()
                .add_xaxis(data_group.index.tolist())
                .add_yaxis('轻度污染', data_group['轻度污染'].tolist(),
                           linestyle_opts=opts.LineStyleOpts(type_='solid',
                                                             color='red'))
                .add_yaxis('中度污染', data_group['中度污染'].tolist(),
                           linestyle_opts=opts.LineStyleOpts(type_='dotted',
                                                             color='green'))
                .add_yaxis('重度污染', data_group['重度污染'].tolist(),
                           linestyle_opts=opts.LineStyleOpts(type_='dashed',
                                                             color='blue'))
                .add_yaxis('严重污染', data_group['严重污染'].tolist(),
                           linestyle_opts=opts.LineStyleOpts(type_='',
                                                             color='orange'))
                .set_series_opts(label_opts=opts.LabelOpts(is_show=False))
                .set_global_opts(
                    title_opts=opts.TitleOpts(title='8月份污染天气变化情况',
                                             pos_left='right',
                                             pos_top='5%'),
                    legend_opts=opts.LegendOpts(pos_left='right', pos_top='10%')
                )
           )
    return line
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30

line().render_notebook()
1

页面组件

page = Page(layout=Page.DraggablePageLayout)

page.add(
    timeline1(),
    timeline2(),
    line()
)
page.render()
1
2
3
4
5
6
7
8

会生成一个html文件，如果复制到浏览器打不开，就从文件资源管理器中打开

打开后，点击Save Config按钮

另存到本地文件后，导入Jupyter Home页面文件夹
重命名保存到Jupyter Home页面文件：

Page.save_resize_html('render.html', cfg_file='chart_config.json', dest='空气可视化.html')
1