pywrangle 0.3.3

Auxiliary functions to clean pandas data frames

Pywrangle

About

PyWrangle is an open-source Python library for data wrangling. Wikipedia defines data wrangling as follows:

is the process of transforming and mapping data from one "raw" data form into another format with the intent of making it more appropriate and valuable for a variety of downstream purposes such as analytics

Functions

PyWrangle currently supports:

• cleaning strings
• tracking dataframe changes
• identifying data entry errors

Documentation & Distribution

Documentation is available here

Distribution is available here

Install

Requirements

• Python >= 3.8
• numpy >= 1.14.4
• pandas >= 1.0.3
• fuzzywuzzy >= 0.18.0
• python-levenshtein >= 0.12.0
• metaphone >= 0.6

Pip Install

To install pywrangle, use pip:

pip install pywrangle

Import

Per convention with Python libraries for data science, import pywrangle as follows:

>>> import pywrangle as pw

Contributing

Like all developers, I love open source. Please reference the contributing guidelines here

History

Version = "0.3.03

• Removed walrus operator for Pre-3.8 compatability. Now Python 3.6+ compatable.

Version = "0.3.0"

• Removed identify missing data from library -- too much overlap with the missingno library.
• Added identify_errors() function. Uses levenshtein's distance & double metaphone string matching algorithms to identify potential data entry errors in string columns.
• Refactored code into different sub libraries
• Placed documentation on ReadTheDocs.

version = "0.2.40"

• refactored code for clarity
• added display info to print_df_changes

version = "0.2.1"

• Created init file for function imports
• Documentation on importing pywrangle
• Added numpy as required package.
• Changed package requirements to greater than or equal to.

version = "0.0.1"

• Init