5：Echarts数据可视化-多条曲线、多个子图、TreeMap类似盒图、树形图、热力图、词云
作者：互联网
〇、目标
本次实验主要基于Echarts的Python库实现高维数据、网络和层次化数据、时空数据和文本数据的可视化，掌握可视化的操作流程和相关库的使用。
一、绘制平行坐标系
平行坐标是信息可视化的一种重要技术。为了克服传统的笛卡尔直角坐标系 难以表达三维以上数据的问题， 平行坐标将高维数据的各个变量用一系列相互平行的坐标轴表示， 变量值对应轴上位置。 为了反映变化趋势和各个变量间相互关系，将描述不同变量的各点连接成折线。
平行坐标因形式的紧凑型和表达的高效性，被广泛使用，但是每个点需要多个像素，数据量大，容易产生视觉混淆。
1、启动Python编辑工具IDLE
双击桌面IDLE图标启动编辑器。
2、新建Python文件
在导航栏中选择File->New File 新建Python文件。
3、编写Python程序
在弹出的新窗口中编写代码如下：
from pyecharts import options as opts
from pyecharts.charts import Page, Parallel
def parallel_base() -> Parallel:
dataBJ = [[1,55,9,56,0.46,18,6,"4"],
[2,25,11,21,0.65,34,9,"5"],
[3,56,7,63,0.3,14,5,"4"],
[4,33,7,29,0.33,16,6,"5"],
[5,42,24,44,0.76,40,16,"5"],
[6,82,58,90,1.77,68,33,"4"],
[7,74,49,77,1.46,48,27,"4"],
[8,78,55,80,1.29,59,29,"4"],
[9,267,216,280,4.8,108,64,"1"],
[10,185,127,216,2.52,61,27,"2"],
[11,39,19,38,0.57,31,15,"5"],
[12,41,11,40,0.43,21,7,"5"],
[13,64,38,74,1.04,46,22,"4"],
[14,108,79,120,1.7,75,41,"3"],
[15,108,63,116,1.48,44,26,"3"],
[16,33,6,29,0.34,13,5,"5"],
[17,94,66,110,1.54,62,31,"4"],
[18,186,142,192,3.88,93,79,"2"],
[19,57,31,54,0.96,32,14,"4"],
[20,22,8,17,0.48,23,10,"5"],
[21,39,15,36,0.61,29,13,"5"],
[22,94,69,114,2.08,73,39,"4"],
[23,99,73,110,2.43,76,48,"4"],
[24,31,12,30,0.5,32,16,"5"],
[25,42,27,43,1,53,22,"5"],
[26,154,117,157,3.05,92,58,"2"],
[27,234,185,230,4.09,123,69,"1"],
[28,160,120,186,2.77,91,50,"2"],
[29,134,96,165,2.76,83,41,"3"],
[30,52,24,60,1.03,50,21,"4"],
[31,46,5,49,0.28,10,6,"5"]];
dataGZ = [[1,26,37,27,1.163,27,13,"5"],
[2,85,62,71,1.195,60,8,"4"],
[3,78,38,74,1.363,37,7,"4"],
[4,21,21,36,0.634,40,9,"5"],
[5,41,42,46,0.915,81,13,"5"],
[6,56,52,69,1.067,92,16,"4"],
[7,64,30,28,0.924,51,2,"4"],
[8,55,48,74,1.236,75,26,"4"],
[9,76,85,113,1.237,114,27,"4"],
[10,91,81,104,1.041,56,40,"4"],
[11,84,39,60,0.964,25,11,"4"],
[12,64,51,101,0.862,58,23,"4"],
[13,70,69,120,1.198,65,36,"4"],
[14,77,105,178,2.549,64,16,"4"],
[15,109,68,87,0.996,74,29,"3"],
[16,73,68,97,0.905,51,34,"4"],
[17,54,27,47,0.592,53,12,"4"],
[18,51,61,97,0.811,65,19,"4"],
[19,91,71,121,1.374,43,18,"4"],
[20,73,102,182,2.787,44,19,"4"],
[21,73,50,76,0.717,31,20,"4"],
[22,84,94,140,2.238,68,18,"4"],
[23,93,77,104,1.165,53,7,"4"],
[24,99,130,227,3.97,55,15,"4"],
[25,146,84,139,1.094,40,17,"3"],
[26,113,108,137,1.481,48,15,"3"],
[27,81,48,62,1.619,26,3,"4"],
[28,56,48,68,1.336,37,9,"4"],
[29,82,92,174,3.29,0,13,"4"],
[30,106,116,188,3.628,101,16,"3"],
[31,118,50,0,1.383,76,11,"3"]];
dataSH = [[1,91,45,125,0.82,34,23,"4"],
[2,65,27,78,0.86,45,29,"4"],
[3,83,60,84,1.09,73,27,"4"],
[4,109,81,121,1.28,68,51,"3"],
[5,106,77,114,1.07,55,51,"3"],
[6,109,81,121,1.28,68,51,"3"],
[7,106,77,114,1.07,55,51,"3"],
[8,89,65,78,0.86,51,26,"4"],
[9,53,33,47,0.64,50,17,"4"],
[10,80,55,80,1.01,75,24,"4"],
[11,117,81,124,1.03,45,24,"3"],
[12,99,71,142,1.1,62,42,"4"],
[13,95,69,130,1.28,74,50,"4"],
[14,116,87,131,1.47,84,40,"3"],
[15,108,80,121,1.3,85,37,"3"],
[16,134,83,167,1.16,57,43,"3"],
[17,79,43,107,1.05,59,37,"4"],
[18,71,46,89,0.86,64,25,"4"],
[19,97,71,113,1.17,88,31,"4"],
[20,84,57,91,0.85,55,31,"4"],
[21,87,63,101,0.9,56,41,"4"],
[22,104,77,119,1.09,73,48,"3"],
[23,87,62,100,1,72,28,"4"],
[24,168,128,172,1.49,97,56,"2"],
[25,65,45,51,0.74,39,17,"4"],
[26,39,24,38,0.61,47,17,"5"],
[27,39,24,39,0.59,50,19,"5"],
[28,93,68,96,1.05,79,29,"4"],
[29,188,143,197,1.66,99,51,"2"],
[30,174,131,174,1.55,108,50,"2"],
[31,187,143,201,1.39,89,53,"2"]];
c = (
Parallel()
.add_schema(
[
{"dim": 0, "name": "date"},
{"dim": 1, "name": "5QI"},
{"dim": 2, "name": "PM2.5"},
{"dim": 3, "name": "PM10"},
{"dim": 4, "name": "3O"},
{"dim": 5, "name": "NO2"},
{"dim": 6, "name": "SO2"},
{"dim": 7, "name": "Level"}
]
)
.add("BJ", dataBJ)
.add("SH", dataSH)
.add("GZ", dataGZ)
.set_global_opts(title_opts=opts.TitleOpts(title="Parallel"))
)
return c
c = parallel_base()
c.render("/home/user/Desktop/parallel.html")
4、保存Python文件
在导航栏中选择File->Save，选择一个文件夹，为文件命名后保存文件。
5、执行Python代码
在菜单栏选择Run->Run Module 执行代码。
6、观察生成图像
代码执行完毕之后，会在桌面上生成一个名为parallel.html的文件，双击文件图标即可在浏览器中观察结果。实验结果如下图：
二、绘制散点图矩阵
散点图矩阵是双变量散点图在多变量情况下的拓展，展现了各个维度两两之间的数据关系。在矩阵中，每一行、每一列均代表一个维度，行与列的维度次序相同。格点中是相应行、列维度所组成的双变量散点图。其中上、下三角矩阵相互对称，可仅展示其中一个三角矩阵以节省显示空间。
1、启动编辑器并新建文件
双击桌面IDLE图标启动IDLE，并在菜单栏选择File->New File 新建Python文件。
2、编写Python程序
编写实验代码如下：
import matplotlib.pyplot as plt
rawData=[[55,9,56,0.46,18,6,"good", "BJ"],
[25,11,21,0.65,34,9,"verygood", "BJ"],
[56,7,63,0.3,14,5,"good", "BJ"],
[33,7,29,0.33,16,6,"verygood", "BJ"],
[42,24,44,0.76,40,16,"verygood", "BJ"],
[82,58,90,1.77,68,33,"good", "BJ"],
[74,49,77,1.46,48,27,"good", "BJ"],
[78,55,80,1.29,59,29,"good", "BJ"],
[267,216,280,4.8,108,64,"severe", "BJ"],
[185,127,216,2.52,61,27,"middle", "BJ"],
[39,19,38,0.57,31,15,"verygood", "BJ"],
[41,11,40,0.43,21,7,"verygood", "BJ"],
[64,38,74,1.04,46,22,"good", "BJ"],
[108,79,120,1.7,75,41,"light", "BJ"],
[108,63,116,1.48,44,26,"light", "BJ"],
[33,6,29,0.34,13,5,"verygood", "BJ"],
[94,66,110,1.54,62,31,"good", "BJ"],
[186,142,192,3.88,93,79,"middle", "BJ"],
[57,31,54,0.96,32,14,"good", "BJ"],
[22,8,17,0.48,23,10,"verygood", "BJ"],
[39,15,36,0.61,29,13,"verygood", "BJ"],
[94,69,114,2.08,73,39,"good", "BJ"],
[99,73,110,2.43,76,48,"good", "BJ"],
[31,12,30,0.5,32,16,"verygood", "BJ"],
[42,27,43,1,53,22,"verygood", "BJ"],
[154,117,157,3.05,92,58,"middle", "BJ"],
[234,185,230,4.09,123,69,"severe", "BJ"],
[160,120,186,2.77,91,50,"middle", "BJ"],
[134,96,165,2.76,83,41,"light", "BJ"],
[52,24,60,1.03,50,21,"good", "BJ"],
[46,5,49,0.28,10,6,"verygood", "BJ"],
[26,37,27,1.163,27,13,"verygood", "GZ"],
[85,62,71,1.195,60,8,"good", "GZ"],
[78,38,74,1.363,37,7,"good", "GZ"],
[21,21,36,0.634,40,9,"verygood", "GZ"],
[41,42,46,0.915,81,13,"verygood", "GZ"],
[56,52,69,1.067,92,16,"good", "GZ"],
[64,30,28,0.924,51,2,"good", "GZ"],
[55,48,74,1.236,75,26,"good", "GZ"],
[76,85,113,1.237,114,27,"good", "GZ"],
[91,81,104,1.041,56,40,"good", "GZ"],
[84,39,60,0.964,25,11,"good", "GZ"],
[64,51,101,0.862,58,23,"good", "GZ"],
[70,69,120,1.198,65,36,"good", "GZ"],
[77,105,178,2.549,64,16,"good", "GZ"],
[109,68,87,0.996,74,29,"light", "GZ"],
[73,68,97,0.905,51,34,"good", "GZ"],
[54,27,47,0.592,53,12,"good", "GZ"],
[51,61,97,0.811,65,19,"good", "GZ"],
[91,71,121,1.374,43,18,"good", "GZ"],
[73,102,182,2.787,44,19,"good", "GZ"],
[73,50,76,0.717,31,20,"good", "GZ"],
[84,94,140,2.238,68,18,"good", "GZ"],
[93,77,104,1.165,53,7,"good", "GZ"],
[99,130,227,3.97,55,15,"good", "GZ"],
[146,84,139,1.094,40,17,"light", "GZ"],
[113,108,137,1.481,48,15,"light", "GZ"],
[81,48,62,1.619,26,3,"good", "GZ"],
[56,48,68,1.336,37,9,"good", "GZ"],
[82,92,174,3.29,0,13,"good", "GZ"],
[106,116,188,3.628,101,16,"light", "GZ"],
[118,50,0,1.383,76,11,"light", "GZ"],
[91,45,125,0.82,34,23,"good", "SH"],
[65,27,78,0.86,45,29,"good", "SH"],
[83,60,84,1.09,73,27,"good", "SH"],
[109,81,121,1.28,68,51,"light", "SH"],
[106,77,114,1.07,55,51,"light", "SH"],
[109,81,121,1.28,68,51,"light", "SH"],
[106,77,114,1.07,55,51,"light", "SH"],
[89,65,78,0.86,51,26,"good", "SH"],
[53,33,47,0.64,50,17,"good", "SH"],
[80,55,80,1.01,75,24,"good", "SH"],
[117,81,124,1.03,45,24,"light", "SH"],
[99,71,142,1.1,62,42,"good", "SH"],
[95,69,130,1.28,74,50,"good", "SH"],
[116,87,131,1.47,84,40,"light", "SH"],
[108,80,121,1.3,85,37,"light", "SH"],
[134,83,167,1.16,57,43,"light", "SH"],
[79,43,107,1.05,59,37,"good", "SH"],
[71,46,89,0.86,64,25,"good", "SH"],
[97,71,113,1.17,88,31,"good", "SH"],
[84,57,91,0.85,55,31,"good", "SH"],
[87,63,101,0.9,56,41,"good", "SH"],
[104,77,119,1.09,73,48,"light", "SH"],
[87,62,100,1,72,28,"good", "SH"],
[168,128,172,1.49,97,56,"middle", "SH"],
[65,45,51,0.74,39,17,"good", "SH"],
[39,24,38,0.61,47,17,"verygood", "SH"],
[39,24,39,0.59,50,19,"verygood", "SH"],
[93,68,96,1.05,79,29,"good", "SH"],
[188,143,197,1.66,99,51,"middle", "SH"],
[174,131,174,1.55,108,50,"middle", "SH"],
[187,143,201,1.39,89,53,"middle", "SH"]];
print(rawData[0:30])
print(rawData[31:62])
print(rawData[63:])
fig = plt.figure()
for i in range(6):
for j in range(6):
if(j<=i):
continue
subfig=fig.add_subplot(5,5,i*5+j)
subfig.scatter([x[i] for x in rawData[0:30]],[x[j] for x in rawData[0:30]],s=5,c='red')
subfig.scatter([x[i] for x in rawData[31:62]],[x[j] for x in rawData[31:62]],s=5,c='green')
subfig.scatter([x[i] for x in rawData[63:]],[x[j] for x in rawData[63:]],s=5,c='blue')
plt.savefig("/home/user/Desktop/scatter.png")
plt.show()
3、保存并执行程序
程序编写完成后，选择菜单栏File->Save 命名并保存文件。保存文件后，选择菜单栏Run->Run Module 执行程序。
4、观察实验结果
代码执行完毕后，会自动显示出实验结果，具体结果如下图，同时会在桌面上产生相对于的png图片
三、绘制TreeMap
TreeMap是一种基于二维空间填充的可视化方法，与传统的层次结构数据可视化方法相比，可以提高屏幕显示空间的利用率，充分利用显示空间的每一个象素，更适合对大型的层次结构数据进行可视化，例如树状的目录结构。
1、启动IDLE编辑器并新建文件
双击桌面IDLE图标启动编辑器，并选择菜单栏File->New File新建文件。
2、编写Python程序
在新建的文件窗口中编写代码如下：
import json
import os
from pyecharts import options as opts
from pyecharts.charts import Page, TreeMap
def treemap_base() -> TreeMap:
with open("/home/user/Data/disk.tree.json","r") as tree_f:
data = json.load(tree_f)
c = (
TreeMap()
.add("present data", data)
.set_global_opts(title_opts=opts.TitleOpts(title="TreeMap"))
)
return c
c = treemap_base()
c.render("/home/user/Desktop/treemap.html")
3、保存并执行代码
选择File->Save保存程序。
保存完成后，选择Run->Run Module执行代码
4、观察实验结果
代码执行结束后，会在桌面生成treemap.html文件，双击文件即可在浏览器中观察结果。
四、绘制树形图
树形图直接将节点之间的父子关系映射到视觉元素中，采用树状结构对层次结构数据进行可视化也是一种非常有效的手段。
1、启动编辑器
新建文件双击桌面IDLE图标，并选择File->New File新建文件
2、编写Python程序
在新建的文件窗口中编辑代码如下：
import json
import os
from pyecharts import options as opts
from pyecharts.charts import Page, Tree
def tree_lr() -> Tree:
with open("/home/user/Data/flare.json") as f:
j = json.load(f)
c = (
Tree()
.add("", [j], collapse_interval=2)
.set_global_opts(title_opts=opts.TitleOpts(title="Tree"))
)
return c
c = tree_lr()
c.render("/home/user/Desktop/tree.html")
3、保存并执行程序
选择File->Save 保存文件
选择Run->Run Module执行代码。
4、观察实验结果
代码执行结束后，会在桌面上生成tree.html文件，双击图标即可在浏览器中观察结果。
五、绘制热力图
1、启动编辑器并新建文件
双击桌面IDLE图标，启动IDLE
选择菜单栏File->New File新建文件
2、编写Python程序
在新建的文件窗口中编写代码如下：
from example.commons import Faker
from pyecharts import options as opts
from pyecharts.charts import Geo
from pyecharts.globals import ChartType, SymbolType
def geo_heatmap() -> Geo:
c = (
Geo()
.add_schema(maptype="china")
.add(
"geo",
[list(z) for z in zip(Faker.provinces, Faker.values())],
type_=ChartType.HEATMAP,
)
.set_series_opts(label_opts=opts.LabelOpts(is_show=False))
.set_global_opts(
visualmap_opts=opts.VisualMapOpts(),
title_opts=opts.TitleOpts(title="Geo-HeatMap"),
)
)
return c
c = geo_heatmap()
c.render("/home/user/Desktop/heatmap.html")
3、保存并执行程序
选择菜单栏File->Save File保存文件
选择菜单栏Run->Run Module执行代码
4、观察实验结果
代码执行完毕后，会在桌面上生成一个名为heatmap.html的文件，双击文件图标即可在浏览器中观察实验结果
六、绘制词云
词云是一种典型的文本可视化技术。该方法把文本数据中的关键词根据出现的频率等规则进行统计，然后进行布局排列，利用颜色、大小等视觉编码其频率信息，该方法可以帮助用户对大规模文本数据有一个概览，已经被广泛用在众多的网站和博客中。
1、启动编辑器并新建文件
双击桌面IDLE图标，启动编辑器。
选择菜单栏File->New File 新建文件
2、编写Python代码
在新建的窗口中编写代码如下：
from pyecharts import options as opts
from pyecharts.charts import Page, WordCloud
from pyecharts.globals import SymbolType
words = [
("C++", 10000),
("C", 6181),
("Java", 4386),
("Python", 4055),
("JavaScript", 2467),
("PHP", 2244),
("XML", 1868),
("Pascal", 1484),
("C#", 1112),
("Object-C", 865),
("HasKell", 847),
("Lisp", 582),
("SQL", 555),
("LabVIEW", 550),
("Logic-based", 462),
("Ada", 366),
("Clojure", 360),
("Ruby", 282),
("Erlang", 273),
("MATLAB", 265),
]
def wordcloud_base() -> WordCloud:
c = (
WordCloud()
.add("", words, word_size_range=[20, 100])
.set_global_opts(title_opts=opts.TitleOpts(title="WordCloud"))
)
return c
c = wordcloud_base()
c.render("/home/user/Desktop/wordcloud.html")
3、保存并执行Python文件
选择File->Save File保存文件。
文件保存结束后，选择Run->Run Module执行程序
4、观察实验结果
程序执行结束后，会在桌面生成wordcloud.html文件，双击文件图标即可在浏览器中观察结果
标签：good,51,盒图,子图,GZ,树形图,SH,BJ,opts 来源： https://www.cnblogs.com/liujinhui/p/16391085.html