Don't have the module building yet, so to install you can run:
pip install .
in the checked out directory.
To uninstall, run:
pip uninstall keydecomposer
This script provides a way to idenfify the smallest unique column set in a pandas dataframe of any size.
pip install pandas
import pandas as pd
import keydecomposer as kd
frame = pd.DataFrame() # generate this however you want
unique_set = kd.decompose_frame(frame)The unique column set is only for the exact dimensions and values of the dataframe passed in. Passing in a partial set of a larger frame will provide a unique column set for the partial frame only.
- Implement a unique column shortcut - instead of parsing the frame, identify a unique column from the start and shortcut the process.
- Add a flag to return all valid unique column sets instead of the smallest one.
- Check and see if the pd.Concat call can be optimized.