CODEWITHSUNDEEP
pythonpandas
Akhil Kintali
Akhil Kintali

Reputation: 546

Python Pandas: Sort an alphanumeric dataframe

I currently have a dataframe as following.

No ID Sub_No Weight
1 a23mcsk 2 30
2 smcd302 3 60
3 a23mcsk 1 24
4 smcd302 2 45
5 a23mcsk 3 18
6 smcd302 1 12

I want to be able to sort this dataframe first by 'ID' and then by the 'Sub_No'. Is there a way I can do this on Python using Pandas?

Expected Result:

No ID Sub_No Weight
3 a23mcsk 1 24
1 a23mcsk 2 30
5 a23mcsk 3 18
6 smcd302 1 12
4 smcd302 2 45
2 smcd302 3 60

Upvotes: 0

Views: 857

Answers (3)

jezrael
jezrael

Reputation: 863031

Use helper column here for correct sorting by numeric values from ID with Sub_No:

df['new'] = df['ID'].str.extract('(\d+)').astype(int)

df = df.sort_values(by=['new', 'Sub_No'], ascending=False).drop('new', axis=1)

Another idea with natural sorting:

import natsort as ns

df['new'] = pd.Categorical(df['ID'], ordered=True, categories= ns.natsorted(df['a'].unique()))
df = df.sort_values(by=['new', 'Sub_No'], ascending=False).drop('new', axis=1)

Upvotes: 1

KaRaOkEy
KaRaOkEy

Reputation: 328

You can use

df.sort_values(by=['ID', 'Sub_No'], ascending=True)

Upvotes: 0

SeaBean
SeaBean

Reputation: 23217

Use:

df.sort_values(by=['ID', 'Sub_No'])

Upvotes: 0

Related Questions