pyjanitor

Common errors

• ModuleNotFoundError: No module named 'pyjanitor'

  cause The pyjanitor package is not installed in the current Python environment or the environment in use is not the one where pyjanitor was installed.
  fix Ensure pyjanitor is installed in your active environment: `pip install pyjanitor`. If using virtual environments, activate the correct environment before running your code.

• AttributeError: 'DataFrame' object has no attribute 'clean_names'

  cause The `janitor` module was not imported, which means its DataFrame accessor methods have not been registered with pandas.
  fix Add `import janitor` to your script after `import pandas as pd`. This registers pyjanitor's functions as DataFrame methods.

• TypeError: 'DataFrameGroupBy' object has no attribute 'mutate' (or similar for other deprecated methods on groupby objects)

  cause Attempting to use a deprecated pyjanitor method on a pandas GroupBy object, or before the relevant methods were added to GroupBy objects.
  fix For operations on grouped DataFrames, use `df.groupby(...).assign(...)` instead of `mutate`. Refer to pyjanitor's documentation for the correct methods available on GroupBy objects for your version.

Warnings

• deprecated The `mutate` DataFrame method has been deprecated. Users are advised to transition to alternative approaches for adding or modifying columns.
  Fix: Use `pd.DataFrame.assign` for general column additions/modifications. For groupby operations, utilize the `assign` method directly on the groupby object: `df.groupby('col').assign(...)` which was introduced in v0.32.18.
• breaking Direct usage of 'by' methods for groupby operations on DataFrames has been migrated to be directly available on groupby objects for improved API consistency.
  Fix: Ensure that `groupby` operations access methods directly on the `DataFrameGroupBy` object returned by `.groupby()`. Review documentation examples for updated patterns.
• deprecated Functions like `add_column`, `add_columns`, `remove_columns`, `rename_column`, `rename_columns`, and `filter_on` are slated for deprecation in a future 1.x release, as their functionality largely overlaps with native pandas methods.
  Fix: Prefer native pandas functions such as `pd.DataFrame.assign`, `pd.DataFrame.drop`, `pd.DataFrame.rename`, and `pd.DataFrame.query` where possible.

Install

• pip install pyjanitor PyPI

Imports

• janitor

  import janitor

  Imports pyjanitor's functionality, registering its methods as pandas DataFrame accessors/methods.
• pandas

  import pandas as pd

  pandas is a prerequisite for pyjanitor's functionality.

Quickstart

This quickstart demonstrates how to install pyjanitor, import it alongside pandas, and use the `clean_names()` function to standardize column headers in a DataFrame for easier manipulation. This function automatically converts names to lowercase and replaces spaces and special characters with underscores.

import pandas as pd
import janitor

# Sample DataFrame with messy column names
data = {
    'First Name': ['Alice', 'Bob'],
    'Last-Name': ['Smith', 'Johnson'],
    'AGE (Years)': [24, 30]
}
df = pd.DataFrame(data)

print("Original DataFrame:\n", df)

# Clean column names using pyjanitor's clean_names()
cleaned_df = df.clean_names()

print("\nCleaned DataFrame:\n", cleaned_df)
print("\nCleaned column names:", cleaned_df.columns.tolist())