Monadic Pipeline Library for Python
This library was created by a team of enthusiastic software developers / data scientists at Microsoft, who wanted to simplify tasks of data processing and creating complex data pipelines. The library is inspired by the following main ideas:
- Using functional approach to data processing (which implies immutability, lazy evaluation, etc.)
- Using pipe module in Python to achieve data pipelines similar to F#.
- Data pipeline uses dictionaries with different fields as base type, new operations would typically enrich data and add
new fields by using
applyfunction. Those dictionaries are similar to monads, and
applyis similar to lift operation on monads. Thus the naming of the library.
Principal developers of mPyPl: