The plt.boxplot() function draws box plots, common methods and detailed meanings

1. Box plot meaning

The box plot is a statistical graph used to statistic data distribution. You can also roughly see whether the data has symmetry, the degree of dispersion and other information. The meaning of the information in the box diagram is as follows:

• The bottom horizontal line indicates the minimum value
• The top horizontal line indicates the maximum value
• Open black circles indicate outliers
• Black solid circles indicate extreme values
• The box is composed of lower quartile, median and upper quartile

  Outliers, also known as outliers, refer to values ​​greater than 1.5 times the interquartile range. Values ​​within 1.5 to 3 times the interquartile range are represented by open circles. The extreme value is one of the outliers.
  An extreme value refers to a value greater than 3 times the interquartile range.

2. Calculation method

First find out the five characteristic values ​​of a set of data, including the minimum and maximum except for outliers, the median, and the two quartiles (upper quartile Q1 And the lower quartile Q3);
median : arrange all values from small to large , if it is an odd number, take the middle value as the median, and then the middle value will not be used when calculating Q1 and Q3 Use ; for even numbers, take the average of the middle two numbers as the median, and continue to use these two numbers when calculating Q1 and Q3 .
Q1 : The median divides all data into two parts, and the part from the minimum to the median is taken as Q1 according to the median method.
Q3 : Take the same method as Q1, take the median from the median to the maximum .
IQR (Interquartile Range) = Q3-Q1 .
All numbers that are not in the interval (Q1-1.5IQR, Q3+1.5IQR) are outliers , the largest remaining value is the maximum value, and the smallest value is the minimum value.
Eigenvalues (from bottom to top): minimum value, Q1, median, Q3, maximum value.
Five values ​​are depicted on a graph, and the five eigenvalues ​​are on a straight line. The minimum value and Q1 are connected, Q1, The median and Q3 are respectively parallel equal length line segments,
Then, connect the two quartiles to form a box.
Finally, connect the two extreme points and the box to form a box diagram, and then click on the outliers.

3. Drawing

3.1 Draw a single box plot

import matplotlib.pyplot as plt
import numpy as np

#生成data数据
np.random.seed(100)
data = np.random.normal(size=(1000,),loc=0,scale=1)

# 绘图
plt.boxplot(data)
plt.show()

3.2 Draw multiple box plots

import matplotlib.pyplot as plt
import numpy as np

np.random.seed(100)
data = np.random.normal(size=(1000,4),loc=0,scale=1)

plt.boxplot(data)

plt.show()

3.3 Actual combat

def plt_box_iamge(df):
    """
    snrr的五个范围为[5,10)、[10,15)、[15,20)、[20,30)、[30-)，按照五个snrr范围计算对应redchi的箱图
    :param df:包含snrr以及redchi的csv数据（dataFrame）。
    :return:
    """
	# 根据snrr范围对redchi进行筛选。
    df1 = df.loc[df['lam_snrr'] >= 5]
    redchi_1 = df1.loc[df1['lam_snrr'] < 10].redchi

    df2 = df.loc[df['lam_snrr'] >= 10]
    redchi_2 = df2.loc[df2['lam_snrr'] < 15].redchi

    df3 = df.loc[df['lam_snrr'] >= 15]
    redchi_3 = df3.loc[df3['lam_snrr'] < 20].redchi

    df4 = df.loc[df['lam_snrr'] >= 20]
    redchi_4 = df4.loc[df4['lam_snrr'] < 30].redchi

    redchi_5 = df.loc[df['lam_snrr'] >= 30].redchi
    # 绘图
    ax = plt.subplot()
    ax.boxplot([redchi_1, redchi_2, redchi_3, redchi_4, redchi_5])
    # 设置轴坐标值刻度的标签
    ax.set_xticklabels(['5<=snrr<10', '10<=snrr<15', '15<=snrr<20', '20<=snrr<30', '30<=snrr'], fontsize=8)
	#	保存图片 
    plt.savefig('./images/box.jpg')
    plt.show()

if __name__ == '__main__':
    df = pd.read_csv('./inputfile/lamost6w_new.csv')
    df_sc = screening(df)  # 筛选数据 (lamost数据应该在正常值范围内，不然因为数值差过大会导致绘制不出图像!)
    plt_box_iamge(df_sc)

3.3 Detailed parameters

plt.boxplot(x,                      # x：指定要绘制箱图的数据
            notch=None,           # notch：是否是凹口的形式展现箱线图，默认非凹口
            sym=None,              # sym：指定异常点的形状，默认为+号显示
            vert=None,              # vert：是否需要将箱线图垂直摆放，默认垂直摆放
            whis=None,             # whis：指定上下须与上下四分位的距离，默认为1.5倍的四分位差
            positions=None,   # positions：指定箱线图的位置，默认为[0,1,2…]
            widths=None,         # widths：指定箱线图的宽度，默认为0.5
            patch_artist=None,        # patch_artist：是否填充箱体的颜色
            meanline=None,             # meanline：是否用线的形式表示均值，默认用点来表示
            showmeans=None,       # showmeans：是否显示均值，默认不显示
            showcaps=None,           # showcaps：是否显示箱线图顶端和末端的两条线，默认显示
            showbox=None,             # showbox：是否显示箱线图的箱体，默认显示
            showfliers=None,          # showfliers：是否显示异常值，默认显示
            boxprops=None,           # boxprops：设置箱体的属性，如边框色，填充色等
            labels=None,                  # labels：为箱线图添加标签，类似于图例的作用
            flierprops=None,          # filerprops：设置异常值的属性，如异常点的形状、大小、填充色等
            medianprops=None,   # medianprops：设置中位数的属性，如线的类型、粗细等
            meanprops=None,       # meanprops：设置均值的属性，如点的大小、颜色等
            capprops=None,           # capprops：设置箱线图顶端和末端线条的属性，如颜色、粗细等
            whiskerprops=None)   # whiskerprops：设置须的属性，如颜色、粗细、线的类型等

3.4 Common methods

import matplotlib.pyplot as plt
import numpy as np

np.random.seed(100)
data = np.random.normal(size=(1000,4),loc=0,scale=1)

ax = plt.subplot()
ax.boxplot(data)                                 # 绘图
ax.set_xlim([0,5])                               # 设置x轴值的范围  rotation=30
# ax.set_xticks()  							      # 自定义x轴的值
ax.set_xlabel("xlabel")                  # 设置x轴的标签
ax.set_xticklabels(['A','B','C','D'],  rotation=30,fontsize=10)   # 设置x轴坐标值的标签 旋转角度 字体大小
ax.set_title("xcy")       					  # 设置图像标题
ax.legend(labels= ['A','B','C','D'],loc='best',)  # 增加图例
ax.text(x=0.2 , y=3.5 , s="test" ,fontsize=12)   # 增加注

plt.show()

Reference:
Baidu Encyclopedia
matplotlib official documentation