Drop Columns in DataFrame by Label Names or by Index Positions

This article will discuss dropping columns in Pandas DataFrame by label names or index positions. Dropping columns can be achieved using methods like drop(), iloc[], loc[], and iterative approaches. Let’s explore these methods with examples.

Drop Columns by Label Names or Index Positions Using drop()

The simplest way to drop columns by their label names is by using the drop() method. This method allows you to specify the column labels you wish to remove.

DataFrame.drop(labels=None, axis=0, index=None, columns=None, inplace=False)

To drop columns, set the axis parameter to 1 or use the columns parameter.

Let's consider an example:

import pandas as pd

# Creating the dataset
data = {
    'Name': ['Rachel', 'Ross', 'Monica', 'Chandler', 'Phoebe', 'Joey'],
    'Age': [30, 32, 31, 33, 34, 31],
    'Profession': ['Fashion Designer', 'Paleontologist', 'Chef', 'Advertiser', 'Musician', 'Actor'],
    'Favorite_Coffee': ['Latte', 'Black Coffee', 'Cappuccino', 'Espresso', 'Herbal Tea', 'Mocha']
}

df = pd.DataFrame(data)

print("Original Dataset:")
display(df)

Output:

Dropping Columns by Label Names using drop()

Let's remove the Profession and Favorite_Coffee columns.

# Dropping columns by label names
df_dropped_labels = df.drop(columns=['Profession', 'Favorite_Coffee'])

print("\nDataset after dropping 'Profession' and 'Favorite_Coffee':")
display(df_dropped_labels)

Output:

Dropping Columns by Index Positions

Using index positions to drop the same columns (Profession and Favorite_Coffee).

# Dropping columns by index positions
columns_to_drop = [2, 3]  # Index positions of 'Profession' and 'Favorite_Coffee'
df_dropped_indices = df.drop(df.columns[columns_to_drop], axis=1)

print("\nDataset after dropping columns by index positions:")
display(df_dropped_indices)

Output:

Drop Columns Using Index Positions with iloc[]

If you don’t know the column labels but know their positions, you can drop columns using index positions.

import pandas as pd

# Creating the dataset
data = {
    'Name': ['Rachel', 'Ross', 'Monica', 'Chandler', 'Phoebe', 'Joey'],
    'Age': [30, 32, 31, 33, 34, 31],
    'Profession': ['Fashion Designer', 'Paleontologist', 'Chef', 'Advertiser', 'Musician', 'Actor'],
    'Favorite_Coffee': ['Latte', 'Black Coffee', 'Cappuccino', 'Espresso', 'Herbal Tea', 'Mocha']
}

df = pd.DataFrame(data)

# Changing the index to 'A', 'B', 'C', etc.
df.index = ['A', 'B', 'C', 'D', 'E', 'F']

print("Original Dataset:")
print(df)

# Drop the 1st and 3rd columns (index 0 and 2)
df_dropped = df.drop(df.columns[[0, 2]], axis=1)
print(df_dropped)

Output:

Original Dataset:
       Name  Age        Profession Favorite_Coffee
A    Rachel   30  Fashion Designer           Latte
B      Ross   32    Paleontologist    Black Coffee
C    Monica   31              Chef      Cappuccino
D  Chandler   33        Advertiser        Espresso
E    Phoebe   34          Musician      Herbal Tea
F      Joey   31             Actor           Mocha

   Age Favorite_Coffee
A   30           Latte
B   32    Black Coffee
C   31      Cappuccino
D   33        Espresso
E   34      Herbal Tea
F   31           Mocha

Drop Columns Using Slicing with loc[]

The loc[] method allows you to drop a range of columns by their label names. Let's take an example to remove column between two labels, we will be using the same dataset that we have used above:

# Drop columns from 'Age' to 'Favorite_Coffee'
df_dropped = df.drop(df.loc[:, 'Age':'Favorite_Coffee'].columns, axis=1)
print(df_dropped)

Output:

        Name
A    Rachel
B      Ross
C    Monica
D  Chandler
E    Phoebe
F      Joey

Drop Columns Iteratively

If you need to remove columns conditionally, you can use an iterative approach.

# Remove any column where the name is 'Age'
for col in df.columns:
    if col == 'Age':
        del df[col]
        
print(df)

Output:

       Name        Profession Favorite_Coffee
A    Rachel  Fashion Designer           Latte
B      Ross    Paleontologist    Black Coffee
C    Monica              Chef      Cappuccino
D  Chandler        Advertiser        Espresso
E    Phoebe          Musician      Herbal Tea
F      Joey             Actor           Mocha

ankthon

Follow

Improve

Article Tags :

• Python
• Python-pandas
• Python pandas-dataFrame

Practice Tags :

• python