Differences Between iloc and loc in Pandas

Introduction

When working with data in Python, Pandas is a library that often comes to the rescue, especially when dealing with large datasets. One of the most common tasks you'll be performing with Pandas is data indexing and selection. This Byte will introduce you to two powerful tools provided by Pandas for this purpose: iloc and loc. Let's get started!

Indexing in Pandas

Pandas provides several methods to index data. Indexing is the process of selecting particular rows and columns of data from a DataFrame. This can be done in Pandas through explicit index and label-based index methods. This Byte will focus on the latter, specifically on the loc and iloc functions.

What is iloc?

iloc is a Pandas function used for index-based selection. This means it indexes based on the integer positions of the rows and columns. For instance, in a DataFrame with n rows, the index of the first row is 0, and the index of the last row is n-1.

Note: iloc stands for "integer location", so it only accepts integers.

Example: Using iloc

Let's create a simple DataFrame and use iloc to select data.

import pandas as pd

# Creating a simple DataFrame
data = {'Name': ['John', 'Anna', 'Peter', 'Linda'],
        'Age': [28, 24, 35, 32],
        'Profession': ['Engineer', 'Doctor', 'Lawyer', 'Writer']}
df = pd.DataFrame(data)

print(df)

This will output:

   Name  Age Profession
0  John   28   Engineer
1  Anna   24     Doctor
2 Peter   35     Lawyer
3 Linda   32     Writer

Let's use iloc to select the first row of this DataFrame:

first_row = df.iloc[0]
print(first_row)

This will output:

Name          John
Age             28
Profession Engineer
Name: 0, dtype: object

Here, df.iloc[0] returned the first row of the DataFrame. Similarly, you can use iloc to select any row or column by its integer index.

What is loc?

loc is another powerful data selection method provided by Pandas. It's works by allowing you to do label-based indexing, which means you select data based on the data's actual label, not its position. It's one of the two primary ways of indexing in Pandas, along with iloc.

Unlike iloc, which uses integer-based indexing, loc uses label-based indexing. This can be a string, or an integer label, but it's not based on the position. It's based on the label itself.

Note: Label-based indexing means that if your DataFrame's index is a list of strings, for example, you'd use those strings to select data, not their position in the DataFrame.

Example: Using loc

Let's look at a simple example of how to use loc to select data. First, we'll create a DataFrame:

import pandas as pd

data = {
    'fruit': ['apple', 'banana', 'cherry', 'date'],
    'color': ['red', 'yellow', 'red', 'brown'],
    'weight': [120, 150, 10, 15]
}
df = pd.DataFrame(data)
df.set_index('fruit', inplace=True)
print(df)

Output:

        color  weight
fruit                
apple     red     120
banana yellow     150
cherry    red      10
date    brown      15

Now, let's use loc to select data:

print(df.loc['banana'])

Output:

color     yellow
weight       150
Name: banana, dtype: object

As you can see, we used loc to select the row for "banana" based on its label.

Differences Between iloc and loc

The primary difference between iloc and loc comes down to label-based vs integer-based indexing. iloc uses integer-based indexing, meaning you select data based on its numerical position in the DataFrame. loc, on the other hand, uses label-based indexing, meaning you select data based on its label.

Another key difference is how they handle slices. With iloc, the end point of a slice is not included, just like with regular Python slicing. But with loc, the end point is included.

Conclusion

In this short Byte, we showed examples of using the loc method in Pandas, saw it in action, and compared it with its couterpart, iloc. These two methods are both useful tools for selecting data in Pandas, but they work in slightly different ways.