使用seaborn绘制变量分布

一位不愿透露姓名的小菜鸡

于 2023-02-01 16:00:11 发布

阅读量142

点赞数

分类专栏： python基础语法文章标签： python 数据分析

本文链接：https://blog.csdn.net/qq_52186952/article/details/128835456

版权

python基础语法专栏收录该内容

1 篇文章 0 订阅

订阅专栏

一、可视化结果

横坐标为长度，纵坐标为密度分布。该表还绘制最大最小以及中间值。

二、代码

##绘制阴性阳性各自长度分布
##d0是阴性长度分布的list,d1是阳性长度分布的list，str为文件名的字符串
def plot_distribution(d0,d1,str):
    my_colors = ["#517664", "#73AA90", "#94DDBC", "#DAB06C", 
             "#DF928E", "#C97973", "#B25F57"]
    # Plot
    f, (a0, a1) = plt.subplots(1, 2, figsize=(24, 12))
    sns.distplot(a=d0, rug=True, hist=False, 
                rug_kws={"color": my_colors[5]},
                kde_kws={"color": my_colors[5], "lw": 5, "alpha": 0.7},
                ax=a0)
    ##绘制第一个图
    a0.set_title("No M", weight="bold", size=20)##标题
    a0.axvline(x=np.mean(d0), ls=":", lw=2, color="black")###绘制竖直辅助线
    a0.text(x=np.mean(d0)+0.5, y=0.000, s="mean: {}".format(np.mean(d0)), size=17, color="black", weight="bold")
    a0.axvline(x=np.min(d0), ls=":", lw=2, color="black")
    a0.text(x=np.min(d0)+0.5, y=0.004, s="min: {}".format(np.min(d0)), size=17, color="black", weight="bold")
    a0.axvline(x=np.max(d0), ls=":", lw=2, color="black")
    a0.text(x=np.max(d0)-10, y=0.008, s="max: {}".format(np.max(d0)), size=17, color="black", weight="bold")

    ##同样代码绘制第二个图
    sns.distplot(a=d1, rug=True, hist=False, 
                rug_kws={"color": my_colors[2]},
                kde_kws={"color": my_colors[2], "lw": 5, "alpha": 0.7},
                ax=a1)
    a1.set_title("M", weight="bold", size=20)
    a1.axvline(x=np.mean(d1), ls=":", lw=2, color="black")
    a1.text(x=np.mean(d1)+0.5, y=0.000, s="mean: {}".format(np.mean(d1)), size=17, color="black", weight="bold")
    a1.axvline(x=np.min(d1), ls=":", lw=2, color="black")
    a1.text(x=np.min(d1)+0.5, y=0.004, s="min: {}".format(np.min(d1)), size=17, color="black", weight="bold")
    a1.axvline(x=np.max(d1), ls=":", lw=2, color="black")
    a1.text(x=np.max(d1)-10, y=0.008, s="max: {}".format(np.max(d1)), size=17, color="black", weight="bold")


    plt.suptitle("{}_Distribution".format(str), weight="bold", size=25)
    sns.despine(right=True, top=True, left=True)##移除坐标轴
    plt.savefig("./{}.png".format(str))##保存
    plt.clf()##清除figure，防止再次调用时figure重叠
    print("0",np.mean(d0),np.min(d0),np.max(d0))
    print("1",np.mean(d1),np.min(d1),np.max(d1))