Deprecated usage of iloc in pandas notebook 2

[Barderus]

I noticed that the tutorial uses .iloc in the pandas section on notebook 2. At the Humble Data workshop at PyCon US 2025, we learned that certain .iloc behaviors are deprecated in pandas 2.2+ and can confuse beginners. It might be worth keeping only .loc instead to keep things easier to understand.
Thanks for the great resource!

[andrewshandy]

Working on it:)

Do you think we want to include something about df.at and df.iat. It seems to be a new convention in addition to df.loc.
Haven't seen this before, saw it in this blog post as well as Pandas 2.3.0 Documentation

[mborus]

Comment on that:

I was worried about this, too, but it seems that the only thing deprecated it using a callable tuple.

Quote "Deprecated since version 2.2.0: Returning a tuple from a callable is deprecated."

See docs at https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.iloc.html

Nothing else is deprecated, the rest stays, so there's no need to confuse beginners.

the main use case for iloc - getting the rows at an index positions - is very useful and should still be taught.