I’ve just started experimenting with the Polars data frame library in Python.
I really like the programmable API it exposes. In fact I am starting an experimental adapter from the data algebra to Polars. When this is complete one can use the data algebra to run the same data transform in Pandas, SQL, or Polars.
Here is my first experiment: de-duping a data frame using window functions: DupRows.ipynb. This depends on an incomplete and not released to PyPi version of the data algebra adapter. It was a very positive initial experiment.