Quoting Pandas documentation (Essential Basic Functionality) regarding Row or Column-wise Function Application:

Arbitrary functions can be applied along the axes of a DataFrame or Panel using the apply() method, which, like the descriptive statistics methods, take an optional axis argument

Taking advantage of this functionality in combination with lamda functions provides a powerful means of implementing advanced row or column based operations.

Example Implementation

Scenario: URL data has been extracted from a database, and I wish to indicate whether each URL points to an internal or external website, or if it points to an internal shared drive.

Solution: In the following statement I create a new column in my data frame called SourceType, determining the appropriate value for each row by applying the function determine_source_type to each URL value.

df['SourceType'] = df.apply(lambda df_row: determine_source_type(str(df_row['URL'])), axis=1)

Where the determine_source_type function is defined as follows:

def determine_source_type(url):
    url = url.upper()

    internal_url_indicators = ('company.com', 'company.internal.com')
    shared_drive_indicators = ('company_drive_1', 'company_drive_2')
    external_url_indicators = ('HTTP', 'WWW', '.COM')

    if any(indicator in url for indicator in internal_url_indicators):
        return 'INTERNAL_URL'
    if any(indicator in url for indicator in shared_drive_indicators):
        return 'SHARED_DRIVE'
    if any(indicator in url for indicator in external_url_indicators):
        return 'EXTERNAL_URL'
    return 'UNKNOWN'