2

Normal syntax for calling a function is func() but I have noticed that loc[] in pandas is without parentheses and still treated as a function. Is loc [] really a function in pandas?

KawaiKx
  • 9,558
  • 19
  • 72
  • 111
  • 1
    Does this answer your question? [Why/How does Pandas use square brackets with .loc and .iloc?](https://stackoverflow.com/questions/46176656/why-how-does-pandas-use-square-brackets-with-loc-and-iloc) – user202729 Feb 04 '21 at 10:01
  • The [question](https://stackoverflow.com/questions/46176656/why-how-does-pandas-use-square-brackets-with-loc-and-iloc) linked by @user202729 (that was closed and not answered) wasn't a duplicate. Now that it was reopened, however, I added an answer there that includes my answer here and goes more into depth on the "why" this syntax is used. @ OP would you consider taking a look at the *other* question and see if this might be considered a duplicate now? – GPhilo Feb 04 '21 at 10:56

3 Answers3

3

Is loc[ ] a function in Pandas?

No. The simplest way to check is:

import pandas as pd
df = pd.DataFrame()
print(df.loc.__class__)

which prints

<class 'pandas.core.indexing._LocIndexer'>

this tells us that df.loc is an instance of a _LocIndexer class. The syntax loc[] derives from the fact that _LocIndexer defines __getitem__ and __setitem__*, which are the methods python calls whenever you use the square brackets syntax.

*Technically, it's its base class _LocationIndexer that defines those methods, I'm simplifying a bit here

GPhilo
  • 18,519
  • 9
  • 63
  • 89
0

LOC[] is a property that allows Pandas to query data within a dataframe in a standard format. Basically you're providing the index of the data you want.

VictrixHominum
  • 1
  • 3
  • *"Basically you're providing the address of the data you want."*.. no, you're not providing any address. At best, you're providing the *index* of the data you want, but that's also not answering the question of "how does this work" – GPhilo Feb 04 '21 at 10:10
0

Create a simple DataFrame df and look at the type of df.loc:

>>> type(df.loc)
pandas.core.indexing._LocIndexer

This means calling loc on a DataFrame returns an object of the type _LocIndexer, so .loc is not a function or method.

Look at what happens when we inspect the type of an actual function / method of a DataFrame:

>>> type(df.rename)
method

So what really happens here, is that you are accessing an attribute of your DataFrame called loc, which contains an object of type _LocIndexer, which itself implements the dunder methods __getitem__. This is a magic method that can be implemented by an object in order to define what happens when the object is indexed with the square bracket notation, just like a list or a dictionary.

sunnytown
  • 1,844
  • 1
  • 6
  • 13