There’s a common warning in pandas about a SettingWithCopyWarning. While the error message covers some of the possible reasons for the error, it doesn’t cover them all. In this post, I’ll show another source of the error and how to fix it.

import os
import pandas as pd

os.getenv('DATA')

'I:\\Data'

shakespeare_path = os.path.join(os.getenv('DATA'), 'shakespeare.csv')

df = pd.read_csv(shakespeare_path)

df

	Name	Year	Category
0	Titus Andronicus	1592	Tragedy
1	The Comedy of Errors	1594	Comedy
2	Richard II	1595	History
3	Romeo and Juliet	1595	Tragedy
4	A Midsummer Night’s Dream	1595	Comedy
5	King John	1596	History
6	Julius Caesar	1599	Tragedy
7	Othello	1604	Tragedy
8	Macbeth	1606	Tragedy

Let’s say you want to make a subset of the data by copying a couple columns.

sdf = df[['Name', 'Year']]

sdf

	Name	Year
0	Titus Andronicus	1592
1	The Comedy of Errors	1594
2	Richard II	1595
3	Romeo and Juliet	1595
4	A Midsummer Night’s Dream	1595
5	King John	1596
6	Julius Caesar	1599
7	Othello	1604
8	Macbeth	1606

Then you want to continue cleaning it up.

sdf.loc[:, 'Year'] = pd.to_datetime(sdf['Year'])

C:\Users\Julius\anaconda3\lib\site-packages\pandas\core\indexing.py:1773: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  self._setitem_single_column(ilocs[0], value, pi)

The error message doesn’t make any sense. You’re already doing what they suggest trying. What you need to do to avoid this is to make a complete copy of the DataFrame using .copy().

sdf2 = df[['Name', 'Year']].copy()

sdf2.loc[:, 'Year'] = pd.to_datetime(sdf2['Year'])

No more error message!