matplotlib 中以数字方式排序的观察次数的数值分类数据的条形图

Question

I have numerical categorical data in a dataframe which I want to bar plot and order in ascending order by value. I am able to do it in seaborn easily, but cannot figure out how to order it in pandas or matplotlib. Here is the code, as I said, the seaborn plot is ordered correct, but the pandas plot is not.

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

df = pd.DataFrame({
        'category':['blue','green','red','blue','green','red','blue','green','red','violet'],
        'income':[15,1,4,5,3,2.5,3.4,4.4,12,1.6],
})
print(df)
df['income_cat'] = np.ceil(df["income"]/1.5)
df["income_cat"].where(df["income_cat"]<5,5.0,inplace=True)
df['income_cat'].value_counts().plot(kind='bar',sort_columns=True)
sns.catplot(x='income_cat',kind='count',data=df)
plt.show()

Also, any other suggestions to make the plot more "meanigful" are welcome.

Answer 1

I assume you want to order the bars by the value of the index. That can be accomplished by using sort_index() :

df['income_cat'].value_counts().sort_index().plot(kind='bar',sort_columns=True)

Answer 2

You can try:

df['income_cat'].value_counts().sort_index().plot(kind='bar')

Result:

Hope this helps.