Fix AttributeError: 'DataFrame' object has no attribute 'ix'

The Pandas module is a popular Python module used for data analysis purposes.

When working with a Pandas DataFrame object, you may encounter an error that says:

AttributeError: 'DataFrame' object has no attribute 'ix'

You get this AttributeError: 'DataFrame' object has no attribute 'ix' because the ix method has been deprecated and removed in the latest Pandas version.

The following article will show you how to fix this error.

How to reproduce the error

For example, imagine you have the following DataFrame in your Python project:

        Age  Score
Nathan   29     72
Lisa     25     88
Mike     22     67

Now you are tasked to get the first two rows of the data.

If you use the ix method as follows:

import pandas as pd

df = pd.DataFrame(
    {"Age": [29, 25, 22], "Score": [72, 88, 67]}, 
    index=["Nathan", "Lisa", "Mike"]
)

print(df.ix[:2])

But the ix method has been deprecated and removed in Pandas version 1.0.0, so running the code above will produce the error below:

Traceback (most recent call last):
  File ...
    print(df.ix[:2])
    ...
    return object.__getattribute__(self, name)
AttributeError: 'DataFrame' object has no attribute 'ix'

To fix this error, you need to replace the ix method with either iloc or loc method.

Let’s see an example of using these methods next!

Replace ix with iloc or loc

The iloc method is used to access a row by its integer index, so you can use it like this:

import pandas as pd

df = pd.DataFrame(
    {"Age": [29, 25, 22], "Score": [72, 88, 67]}, 
    index=["Nathan", "Lisa", "Mike"]
)

print(df.iloc[:2])

# Output:
#         Age  Score
# Nathan   29     72
# Lisa     25     88

On the other hand, the loc method is used for label-based indexing.

Because the example data has the index parameter, you can use the value for label-based slicing as follows:

import pandas as pd

df = pd.DataFrame(
    {"Age": [29, 25, 22], "Score": [72, 88, 67]}, 
    index=["Nathan", "Lisa", "Mike"]
)

print(df.loc["Lisa"])

# Output
# Age      25
# Score    88
# Name: Lisa, dtype: int64

Keep in mind that when you pass the index parameter to the iloc method, Python will show another error.

Previously, the ix method can process both label-based and integer-based indexing. It will first process label-based indexing, then fall back to integer-based indexing when it fails.

But because the ix method wasn’t consistent, it was deprecated in favor of loc and iloc methods for slicing and indexing.

If you still want to use the ix method for indexing, then you need to downgrade your pandas module to below version 1.0.0.

Conclusion

Now you’ve learned why the AttributeError: 'DataFrame' object has no attribute 'ix' occurs and how to fix it.

To avoid this error, you need to replace the call to ix method with either iloc or loc method.

Good work solving the error. I look forward to seeing you next time! 👋