利用pyecharts读取csv并进行数据统计可视化的实现

因为需要一个html形式的数据统计界面，所以做了一个基于pyecharts包的可视化程序，当然matplotlib还是常用的数据可视化包，只不过各有优劣；基本功能概述就是读取csv文件数据，对每列进行数据统计并可视化，最后形成html动态界面，选择pyecharts的最主要原因就是这个动态界面简直非常炫酷。

先上成品图：

数据读取和数据分析模块：

#导入csv模块
import csv
#导入可视化模块
from matplotlib import pyplot as plt
from pylab import mpl
import numpy as np
import random
from pyecharts import Line,Pie,Grid,Bar,WordCloud
#指定文件名，然后使用 with open() as 打开

python_file = 'haiyang.csv'
#filename = 'release/111.csv'
#python3 LieCharts.py test_chart --python_file 'haiyang.csv'
with open(python_file) as f:
   #创建一个阅读器：将f传给csv.reader
   reader = csv.reader(f)
   #使用csv的next函数，将reader传给next，将返回文件的下一行
   header_row = next(reader)

for index, column_header in enumerate(header_row):
       print(index, column_header)

#读取置信度
   #创建置信度的列表
   confidences =[]
   #创建风险等级数组
   highRisk = []
   middleRisk = []
   lowRisk = []
   noRisk = []
   person = []
   #创建时间点
   timePoint = []
   #文件信息
   fileInformation = []

#遍历reader的余下的所有行（next读取了第一行，reader每次读取后将返回下一行）
   for row in reader:

# 下面就是对某一列数据进行遍历，因为项目保密，就不列出具体代码了，其实就是各种循环语句，大家根据自己的数据简单写一下就行

fileInformation.append('某某某某')
   fileInformation.append(row[0])
   fileInformation.append(row[1])
   fileInformation.append(row[2])
   fileInformation.append(len(confidences))
   int_confidences = []
   for i in confidences:
 # 同上
   len_noRisk = len(noRisk)
   len_lowRisk = len(lowRisk)
   len_middleRisk = len(middleRisk)
   len_highRisk = len(highRisk)
   len_person = len(person)

total = int(len_person+len_highRisk+len_middleRisk+len_lowRisk+len_noRisk)
   if (len_highRisk > total/2):
 # 同上

数据可视化模块：

pie_title = Pie('某某某分析报表', "", title_pos='center',title_top="1％",title_text_size=42,subtitle_text_size=20)

value=[10000,6181,4386,4055,4000]
wordcloud=WordCloud(width=30,height=12,title="某某某某信息",title_pos="22％",title_top="12％",title_text_size=32)
wordcloud1=WordCloud(width=30,height=12,title="某某："+fileInformation[1],title_pos="22％",title_top="22％",title_text_size=26)
wordcloud2=WordCloud(width=30,height=12,title="某某："+fileInformation[2],title_pos="22％",title_top="30％",title_text_size=26)
#wordcloud3=WordCloud(width=30,height=12,title="音频采样率："+fileInformation[3],title_pos="22％",title_top="38％",title_text_size=26)
#wordcloud4=WordCloud(width=30,height=12,title="总时长/s："+fileInformation[4],title_pos="22％",title_top="36％",title_text_size=32)

# wordcloud.add("",fileInformation,value,word_size_range=[20,100],rotate_step=3
#        ,xaxis_pos=200,grid_left="1％",grid_bottom="50％",grid_top="5％",grid_right="80％")
#折线图
line=Line("某某某某某走势图",title_pos='center',title_top="51％",title_text_size=32,width=600,height = 20)
attr=timePoint
line.add("某某某某某",attr,int_confidences,legend_pos="85％",legend_top="54％",
   mark_point=["max","min"],mark_line=["average"])
#饼图
attr=["某某某某", "某某某某", "某某某某", "某某某"]
v1=[len_highRisk, len_middleRisk, len_lowRisk,len_noRisk]
pie=Pie("某某某某某某某",title_pos="65％",title_top="12％",title_text_size=32,width=100,height = 100)
pie.add("",attr,v1,radius=[0,30],center=[71,35],
   legend_pos="85％",legend_top="20％" ,legend_orient="vertical")
grid=Grid(width = 1800 ,height= 900)#调整画布大小

grid.add(line,grid_left="5％",grid_bottom="2％",grid_top="60％")
grid.add(pie_title,grid_bottom="10％")
grid.add(wordcloud,grid_left="1％",grid_bottom="50％",grid_top="5％",grid_right="80％")
grid.add(wordcloud1,grid_left="1％",grid_bottom="50％",grid_top="5％",grid_right="80％")
grid.add(wordcloud2,grid_left="1％",grid_bottom="50％",grid_top="5％",grid_right="80％")
#grid.add(wordcloud3,grid_left="1％",grid_bottom="50％",grid_top="5％",grid_right="80％")
#grid.add(wordcloud4,grid_left="1％",grid_bottom="50％",grid_top="5％",grid_right="80％")
grid.add(pie,grid_left="50％",grid_bottom="50％")

#grid.render()
grid.render(path='./release/XXXX.html')

根据需求这个还可以跨平台跨语言调用，比如C++程序调用python进行数据分析。

来源：https://juejin.im/post/5e97e5b66fb9a03c5c5a98e4