How to select data from pandas DataFrames with loc[]

In the Python pandas library, DataFrame.loc[] is a property that lets you select data from a DataFrame using labels. This makes it easy to extract specific rows and columns from a DataFrame.

What is the syntax for pandas loc[]?

The syntax for loc[] is quite simple. All you need to do is pass the labels of the columns and rows you want to select as a parameter:

DataFrame.loc[selection]

With pandas loc[], selections are primarily made using labels. This means the parameter you provide can be a single label, a list or a slice of labels. Boolean arrays can also be used as well.

What is the difference between loc[] and iloc[]?

While pandas DataFrame.loc[] selects data based on labels, DataFrame.iloc selects data based on integer-based positions. Here’s a code example to help illustrate the differences. First, we’re going to create a pandas DataFrame:

import pandas as pd
# Example DataFrame
data = {'Name': ['Alyssa', 'Brandon', 'Carmen'], 'Age': [23, 35, 30]}
df = pd.DataFrame(data)
print(df)

Here’s what the DataFrame looks like:

Name    Age
0   Alyssa     	23
1 	Brandon     35
2  	Carmen     	30

To extract “Alyssa” from the DataFrame, you can use both pandas loc[] and iloc[]. Although the approach differs, the result is the same:

# Using loc and labels to extract Alyssa
print(df.loc[0, 'Name'])  # Output: 'Alyssa'
# Using iloc and integers to extract Alysa
print(df.iloc[0, 0])  # Output: 'Alyssa'

How to use pandas DataFrame.loc[]

Pandas loc[] helps you extract subsets of your DataFrame. With loc[], you can extract a single row or column, multiple rows and columns or even apply conditions for filtering. This flexibility makes it suitable for a variety of use cases.

Selecting a single row

Let’s look at a DataFrame example:

import pandas as pd
data = {
    'Name': ['Alyssa', 'Brandon', 'Carmen'],
    'Age': [23, 35, 30],
    'City': ['Detroit', 'Atlanta', 'Seattle']
}
df = pd.DataFrame(data)
print(df)

Here’s what the resulting DataFrame looks like:

Name  	Age     City
0   Alyssa    23  	Detroit
1 	Brandon   35    Atlanta
2 	Carmen    30    Seattle

To select the data from the row that contains information about Brandon (index 1), you can use pandas loc[]:

brandon_data = df.loc[1]
print(brandon_data)

Here’s the result:

Name         Brandon
Age              35
City        	Atlanta
Name: 1, dtype: object

Selecting multiple columns

You can also use DataFrame.loc[] to select a subset of columns. The following code selects the columns “Name” and “City”:

name_city = df.loc[:, ['Name', 'City']]
print(name_city)

The result is a subset of the original DataFrame:

Name     City
0   Alyssa  Detroit
1 Brandon   Atlanta
2  Carmen   Seattle

Selecting rows based on conditions

With pandas loc[], you can also select rows that meet specific criteria. You can do this with Boolean comparison operators. For example, here’s how to filter out all individuals who are older than 25:

older_than_25 = df.loc[df['Age'] > 25]
print(older_than_25)

The code above produces a DataFrame that only includes data for individuals in the DataFrame who are older than 25:

Name  	Age     City
1 Brandon     35   Atlanta
2  Carmen     30   Seattle