Box plot rendering

insert image description here
This is a boxplot made using python, showing 10 boxplots in the same axis. Boxplots are great for showing discrete distributions of data. In addition to the picture display, python can also calculate the relevant data of each boxplot, such as the average value, maximum value, minimum value, etc. of the target data, for example (some data corresponding to the above boxplot):

          index1      index2      index3     
count  842.000000  842.000000  842.000000    
mean     9.284371    9.044644    8.989394   
std      0.240151    0.261572    0.267525   
min      7.880000    7.800000    7.300000   
25%      9.162500    8.900000    8.840000   
50%      9.320000    9.070000    9.020000    
75%      9.440000    9.217500    9.160000       
max      9.800000    9.760000    9.730000

the code

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import xlrd

#保存数据，可用于之后的学习研究
def save_data(data, path):
    Note=open(path,mode='w')
    print("保存数据")
    Note.write(str(data))
    Note.close()

#读取xls文件数据
def read_excel(excel_name,sheet_name):
    """参数：
    输入：excel的名字
    输出：目标表的名字"""
    workbook = xlrd.open_workbook(excel_name)  #打开文件
    sheet = workbook.sheet_by_name(sheet_name)  #读取sheet页
    # 获取表的行列数
    rows = sheet.nrows
    cols = sheet.ncols
    all_data = []
    tempList = []
    # 读取放置在excel中的数据
    for col in range(3, 13):
        for row in range(1, rows):
            print(rows)
            tempList.append(sheet.cell(row, col).value)
        all_data.append(tempList)
        tempList = []
    return all_data

#设置盒图的图片参数
def draw_box(all_data):
    for i in range(1):
        db = np.array(all_data)  # 把数据放置到numpy中的数组中
        print(db)
        fig = plt.figure(figsize=(40, 15))  # 设置画布大小
        ax = fig.add_subplot(111)  # 图形在画布中的布局
        allg = ["index1", "index2", "index3", "index4", "index5", "index6", "index7", "index8", "index9","index10"]  # 设置横坐标
        df = pd.DataFrame(db.T, columns=allg)  # db.T表示转置
        ax.set_title("Box-and-Whisker Plots", fontsize=24)  # 以下表示设置坐标标签以及字体大小
        ax.set_xlabel('index', fontsize=22)
        ax.set_ylabel('grade', fontsize=22)
        df.boxplot(ax=ax, fontsize=24)
        plt.savefig("盒图4.jpg")  # 保存
        plt.show()
        pd.set_option('display.max_columns', None)  # 为了将数据全部展示，显示全部列
        data = df.describe()  # 计算相关指标
        print(data)
        save_data(data, "数据.txt")  # 用于存储各组数据的指标数据


if __name__ == '__main__':
    all_data = read_excel("参数1", "参数2")#读取数据，参数1是excel的名字，参数2是里面具体的表名
    draw_box(all_data)#绘制图表

Instructions

data

The data used in this code is in units of columns, that is, each column of data corresponds to a box plot.
In the for loop in the read_excel() function when reading the table data:

for col in range(3, 13):
        for row in range(1, rows):

(The subscript of the first column in the excel table starts from 0), you can modify it according to your needs.

chart

allg = ["index1", "index2", "index3", "index4", "index5", "index6", "index7", "index8", "index9","index10"]  # 设置横坐标

allg sets the abscissa, which can be modified according to your actual data

The following is my git address, welcome everyone to download for free, exchange and learn:

https://github.com/HYHJessica/

[python] box plot production