如何在Pandas DataFrame中按给定列表排序列?

Question

给定这个DataFrame和一个列表:

df = pd.DataFrame([['August', 2], ['July', 3], ['Sept', 6]], columns=['A', 'B'])
order = ['July','Sept','August']
df
Out[310]: 
A  B
1  August  2
0    July  3
2    Sept  6

我想用给定的列表重新排序df['A']，所以最终输出是:

A  B
1    July  3
0    Sept  6
2  August  2

是否有任何方法，使我可以排序的列使用字符串列表?

Answer 1

如果可能的话，按前3个字母生成的日期时间排序使用DataFrame.sort_values和key参数:

df = df.sort_values('A', key=lambda x: pd.to_datetime(x.str[:3], format='%b'))
print (df)
A  B
1    July  3
0  August  2
2    Sept  6

另一个有序分类的解:

order = ['July','Sept','August']
df['A'] = pd.Categorical(df.A, ordered=True, categories=order)
df = df.sort_values('A')
print (df)
A  B
1    July  3
2    Sept  6
0  August  2

Answer 2

试一试:

df = pd.DataFrame([['August', 2], ['July', 3], ['Sept', 6]], columns=['A', 'B'])
order = ['July','Sept','August']
df = df.set_index('A')
df = df.reindex(index = order)
df = df.reset_index()