如何在Seaborn关节图上手动更改边距图的刻度标签 [英] How to manually change the tick labels of the margin plots on a Seaborn jointplot

问题描述

我正在尝试使用对数标度作为我的海底联合图的边距图.我正在使用set_xticks()和set_yticks()，但是我的更改没有出现.这是下面的代码以及生成的图形:

I am trying to use a log scale as the margin plots for my seaborn jointplot. I am usings set_xticks() and set_yticks(), but my changes do not appear. Here is my code below and the resulting graph:

import matplotlib.pyplot as plt
%matplotlib inline
import numpy as np
import seaborn as sns
import pandas as pd

tips = sns.load_dataset('tips')
female_waiters = tips[tips['sex']=='Female']

def graph_joint_histograms(df1):
    g=sns.jointplot(x = 'total_bill',y = 'tip', data = tips, space = 0.3,ratio = 3)
    g.ax_joint.cla()
    g.ax_marg_x.cla()
    g.ax_marg_y.cla()

    for xlabel_i in g.ax_marg_x.get_xticklabels():
        xlabel_i.set_visible(False)
    for ylabel_i in g.ax_marg_y.get_yticklabels():
        ylabel_i.set_visible(False)

    x_labels = g.ax_joint.get_xticklabels()
    x_labels[0].set_visible(False)
    x_labels[-1].set_visible(False)

    y_labels = g.ax_joint.get_yticklabels()
    y_labels[0].set_visible(False)
    y_labels[-1].set_visible(False)

    g.ax_joint.set_xlim(0,200)
    g.ax_marg_x.set_xlim(0,200)

    g.ax_joint.scatter(x = df1['total_bill'],y = df1['tip'],data = df1,c = 'y',edgecolors= '#080808',zorder = 2)
    g.ax_joint.scatter(x = tips['total_bill'],y = tips['tip'],data = tips, c= 'c',edgecolors= '#080808')

    ax1 =g.ax_marg_x.get_axes()
    ax2 = g.ax_marg_y.get_axes()
    ax1.set_yscale('log')
    ax2.set_xscale('log')

    ax1.set_yscale('log')
    ax2.set_xscale('log')

    ax2.set_xlim(1e0, 1e4)
    ax1.set_ylim(1e0, 1e3)
    ax2.xaxis.set_ticks([1e0,1e1,1e2,1e3])
    ax2.xaxis.set_ticklabels(("1","10","100","1000"), visible = True)


    plt.setp(ax2.get_xticklabels(), visible = True)
    colors = ['y','c']
    ax1.hist([df1['total_bill'],tips['total_bill']],bins = 10, stacked=True,log = True,color = colors, ec='black')

    ax2.hist([df1['tip'],tips['tip']],bins = 10,orientation = 'horizontal', stacked=True,log = True,color = colors, ec='black')
ax2.set_ylabel('')

任何想法都将不胜感激.

Any ideas would be much appreciated.

这是结果图:

推荐答案

由于轴没有get_axes()方法，因此您实际上应该从g.ax_marg_y.get_axes()行得到一个错误. 对此进行纠正

You should actually get an error from the line g.ax_marg_y.get_axes() since an axes does not have a get_axes() method. Correcting for that

ax1 =g.ax_marg_x
ax2 = g.ax_marg_y

应该给你想要的情节.不幸的是，对数轴的刻度标签被直方图的log=True参数覆盖.因此，您可以省去该设置(因为您已经设置了轴无论如何都可以对数缩放)，或者您需要在调用hist的之后设置标签.

should give you the desired plot. The ticklabels for the log axis are unfortunately overwritten by the histogram's log=True argument. So you can either leave that out (since you already set the axes to log scale anyways) or you need to set the labels after calling hist.

import matplotlib.pyplot as plt
import seaborn as sns

tips = sns.load_dataset('tips')

def graph_joint_histograms(tips):
    g=sns.jointplot(x = 'total_bill',y = 'tip', data = tips, space = 0.3,ratio = 3)
    g.ax_joint.cla()
    g.ax_marg_x.cla()
    g.ax_marg_y.cla()

    for xlabel_i in g.ax_marg_x.get_xticklabels():
        xlabel_i.set_visible(False)
    for ylabel_i in g.ax_marg_y.get_yticklabels():
        ylabel_i.set_visible(False)

    x_labels = g.ax_joint.get_xticklabels()
    x_labels[0].set_visible(False)
    x_labels[-1].set_visible(False)

    y_labels = g.ax_joint.get_yticklabels()
    y_labels[0].set_visible(False)
    y_labels[-1].set_visible(False)

    g.ax_joint.set_xlim(0,200)
    g.ax_marg_x.set_xlim(0,200)

    g.ax_joint.scatter(x = tips['total_bill'],y = tips['tip'],data = tips,
                       c = 'y',edgecolors= '#080808',zorder = 2)
    g.ax_joint.scatter(x = tips['total_bill'],y = tips['tip'],data = tips, 
                       c= 'c',edgecolors= '#080808')

    ax1 =g.ax_marg_x
    ax2 = g.ax_marg_y
    ax1.set_yscale('log')
    ax2.set_xscale('log')

    ax2.set_xlim(1e0, 1e4)
    ax1.set_ylim(1e0, 1e3)

    ax2.xaxis.set_ticks([1e0,1e1,1e2,1e3])
    ax2.xaxis.set_ticklabels(("1","10","100","1000"), visible = True)

    plt.setp(ax2.get_xticklabels(), visible = True)
    colors = ['y','c']
    ax1.hist([tips['total_bill'],tips['total_bill']],bins = 10, 
             stacked=True, color = colors, ec='black')

    ax2.hist([tips['tip'],tips['tip']],bins = 10,orientation = 'horizontal', 
             stacked=True, color = colors, ec='black')
    ax2.set_ylabel('')


graph_joint_histograms(tips)
plt.show()

这篇关于如何在Seaborn关节图上手动更改边距图的刻度标签的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！