使用Seaborn反转色调类别的顺序和颜色

Question

所以，当我绘制一个方框图，并且在“色调”中使用一个cotagory时，我会得到一个默认的类别顺序（在我的例子中是和否）。我希望盒子以相反的顺序（否是）和相反的颜色呈现。这是我的代码：

fig=plt.figure(figsize=(12,6))
df2=df[df["AMT_INCOME_TOTAL"]<=df["AMT_INCOME_TOTAL"].median()]
ax = sns.boxplot(x="NAME_FAMILY_STATUS", y="AMT_INCOME_TOTAL", data=df2,hue="FLAG_OWN_REALTY")
plt.legend(bbox_to_anchor=(1.05,1),loc=2, borderaxespad=1,title="Own Property flag")
plt.title('Income Totals per Family Status for Bottom Half of Earners')

我想先用蓝色的“N”类

Answer 1

如上所述，如果要明确指定框的顺序和色调顺序，则此操作将执行：

fig=plt.figure(figsize=(12,6))
df2=df[df["AMT_INCOME_TOTAL"]<=df["AMT_INCOME_TOTAL"].median()]
ax = sns.boxplot(x="NAME_FAMILY_STATUS", y="AMT_INCOME_TOTAL", data=df2,hue="FLAG_OWN_REALTY",order=['Married',"Civil marriage","Widow","Single / not married","Separated"], hue_order = ['N', 'Y'])
plt.legend(bbox_to_anchor=(1.05,1),loc=2, borderaxespad=1,title="Own Property flag")
plt.title('Income Totals per Family Status for Bottom Half of Earners')

Answer 2

只需在sns.boxplot（）上添加色调顺序

ax = sns.boxplot(x="NAME_FAMILY_STATUS", y="AMT_INCOME_TOTAL", data=df2,hue="FLAG_OWN_REALTY", hue_order = ['N', 'Y'])