使用 seaborn 绘制条形图

Question

我正在使用 seaborn 绘制以下值：

import pandas as pd
f = [0.31,0.75,0.75,0.75, 0.66,0.83,0.57, 0.69, 0.64,0.61, 0.81,0.21,0.71,0.71,0.64,0.55,0.72,0.74,0.73,0.77]
p = [0.53, 0.72,0.73,0.70,0.44,0.38,0.68,0.79,0.60,0.76, 0.76,0.32,0.84,0.79,0.80,0.38,0.77,0.86,0.81,0.79]
r = [0.74,0.54,0.63,0.61,0.41,0.83,0.63,0.67,0.63,0.53, 0.86,0.51,0.21,0.68,0.59,0.98,0.78,0.75,0.71,0.67]
a = [0.92,0.93,0.92,0.92,0.94,0.82,0.97,0.94,0.91,0.93, 0.97,0.91,0.94,0.93,0.94,0.71,0.93,0.91,0.85,0.94]
pp=['B','B','B','B','B','T','T','T','T','T','B','B','B','B','B','T','T','T','T','T']  
m=['N','L','S','G','Rt','N','L','S','G','Rt','N','L','S','G','Rt','N','L','S','G','Rt']
d=['yes','yes','yes','yes','yes','yes','yes','yes','yes','yes','no','no','no','no','no','no','no','no','no','no']

df = {'DD':d, 'PP': pp, 'M': m, 'P': p, 'R':r, 'F':f,'A':a}
df2= pd.DataFrame(data=df)
df2 = df.melt(['PP', 'M', 'D'])

情节的代码是

g = sns.catplot(
    data=df2,
    x='m',
    y='value',
    hue='PP',
    col='variable',
    col_wrap=2,
    col_order = ['P', 'R', 'F', 'A'],
    kind='bar',
     ci=None,
    facet_kws={'sharey': False, 'sharex': False},
height=8.)

d_col = df2['DD'][:8]
for ax in g.axes.flat:
    labels = ax.get_xticklabels()
    for i,l in enumerate(labels):
        tmp = l.get_text()
        labels[i] = tmp + '\n' + d_col[i]
        ax.set_xticklabels(labels)
sns.set_style(style='white')

然而情节似乎是错误的。另外，我看不到no 值，而只能看到yes。我的预期输出是

在绘图示例中，我为条形图使用了虚拟值。每个 m 应该同时包含 yes 和 no，并且每个子图都将重复此操作（仅出于示例目的留空）。

您能告诉我如何生成类似的输出吗？谢谢

Answer 1

您正在为每个“M”创建两个条（一个用于“PP”为“B”，一个用于“T”）。每个“DD”没有分隔条。 Seaborn 仅接受 hue 的一列。您可以将 'PP' 和 'DD' 列组合成一个新列并将其用于hue，从而为每个 'M' 生成 4 个条形。

这是一个代码示例。此外，sns.catplot 似乎希望将 sharey 和 sharex 直接作为关键字（至少在当前版本 0.11.1 中）。

from matplotlib import pyplot as plt
import seaborn as sns
import pandas as pd
import numpy as np

f = [0.31,0.75,0.75,0.75,0.66,0.83,0.57,0.69,0.64,0.61,0.81,0.21,0.71,0.71,0.64,0.55,0.72,0.74,0.73,0.77]
p = [0.53,0.72,0.73,0.70,0.44,0.38,0.68,0.79,0.60,0.76,0.76,0.32,0.84,0.79,0.80,0.38,0.77,0.86,0.81,0.79]
r = [0.74,0.54,0.63,0.61,0.41,0.83,0.63,0.67,0.63,0.53,0.86,0.51,0.21,0.68,0.59,0.98,0.78,0.75,0.71,0.67]
a = [0.92,0.93,0.92,0.92,0.94,0.82,0.97,0.94,0.91,0.93,0.97,0.91,0.94,0.93,0.94,0.71,0.93,0.91,0.85,0.94]
pp = ['B','B','B','B','B','T','T','T','T','T','B','B','B','B','B','T','T','T','T','T']
m = ['N','L','S','G','Rt','N','L','S','G','Rt','N','L','S','G','Rt','N','L','S','G','Rt']
d = ['yes','yes','yes','yes','yes','yes','yes','yes','yes','yes','no','no','no','no','no','no','no','no','no','no']

df = {'DD': d, 'PP': pp, 'M': m, 'P': p, 'R': r, 'F': f, 'A': a}
df1 = pd.DataFrame(data=df)
df2 = df1.melt(['PP', 'M', 'DD'])
df2['PP_DD'] = df2['PP'] + ', ' + df2['DD']

sns.set_style(style='white')
g = sns.catplot(
    data=df2,
    x='M',
    y='value',
    hue='PP_DD',
    col='variable',
    col_wrap=2,
    col_order=['P', 'R', 'F', 'A'],
    kind='bar',
    ci=None,
    sharey=False, sharex=False,
    height=3, aspect=1.5)
g.fig.subplots_adjust(left=0.06, bottom=0.09)
plt.show()