python 一张图中绘制一个小的子图

有时候为了直观展现图的信息，可以在大图中添加小子图的方式进行数据分析，如下图所示：

具体的代码如下：该图连接了数据库，当然重要的不是数据展示，而是添加子图的方法。

import matplotlib.pyplot as plt
import MySQLdb as mdb
import numpy as np
from mpl_toolkits.axes_grid1.inset_locator import inset_axes
from mpl_toolkits.axes_grid1.inset_locator import mark_inset


def graph():
    # 连接数据库
    conn = mdb.connect(host='127.0.0.1', port=3306, user='root', passwd='root', db='alibaba_trace', charset='utf8')

    # 如果使用事务引擎，可以设置自动提交事务，或者在每次操作完成后手动提交事务conn.commit()
    conn.autocommit(1)  # conn.autocommit(True)

    # 使用cursor()方法获取操作游标
    cursor = conn.cursor()
    # 因该模块底层其实是调用CAPI的，所以，需要先得到当前指向数据库的指针。
    try:
        cursor.execute("select machineID, count(id) from batch_instance where machineID != 0 group by machineID")
        records = cursor.fetchall()
        list_records = list(records)

    except:
        import traceback
        traceback.print_exc()
        # 发生错误时回滚
        conn.rollback()
    finally:
        # 关闭游标连接
        cursor.close()
        # 关闭数据库连接
        conn.close()

    res = []
    res[:] = map(list, list_records)
    machineID = [x[0] for x in res]
    instance_num = [x[1] for x in res]
    print(max(instance_num))
    print(min(instance_num))


    fig = plt.figure()
    ax1 = fig.add_subplot(1, 1, 1)
    # # cdf
    # hist, bin_edges = np.histogram(instance_num, bins=len(np.unique(instance_num)))
    # cdf = np.cumsum(hist / sum(hist))
    # ax1.plot(bin_edges[1:], cdf, color='red', ls='-')
    # ax1.set_xlabel("instance number per machine")
    # ax1.set_ylabel("portion of machine")
    # plt.savefig('../../imgs_mysql/cdf_of_machine_instance.png')

    # # 直方图
    ax1.hist(instance_num, normed=False, alpha=1.0, bins=100)
    ax1.set_xlabel('instance number per machine')
    ax1.set_ylabel('machine number')
    # cdf  要添加的子图
    axins = inset_axes(ax1, width=1.5, height=1.5, loc='upper left')
    # ax1 大图
    # width height分别为子图的宽和高
    # loc 为子图在大图ax1中的相对位置 相应的值有
    # upper left
    # lower left
    # lower right
    # right
    # center left
    # center right
    # lower center
    # upper center
    # center
    hist, bin_edges = np.histogram(instance_num, bins=len(np.unique(instance_num)))
    cdf = np.cumsum(hist / sum(hist))
    axins.plot(bin_edges[1:], cdf, color='red', ls='-')
    axins.set_yticks([])
    # axins.set_xlabel("instance number per machine")
    # axins.set_ylabel("portion of machine")

    plt.savefig("../../imgs_mysql/hist_of_machine_instance")
    plt.show()

if __name__ == '__main__':
    graph()