ashwinmoorkoth1 / pidd-iteratively-imputing-missing-data

This jupyter notebook was developed to be a tutorial on how to handle missing data using scikit-learn's IterativeImputer and shows the differences in predictions made after modeling on the iteratively imputed datasets vs datasets with imputations of other means.

PIDD: Iteratively Imputing Missing Data

This jupyter notebook was developed to be a tutorial on how to handle missing data using scikit-learn's IterativeImputer and shows the differences in predictions made after modeling on the iteratively imputed datasets vs datasets with imputations of other means. The dataset used for this notebook is PIMA Indian's Diabetes Database which can be found on Kaggle for free.
Note: I've included the dataset here as well, as a csv file.
I've created multiple functions so that users can test out their own estimator-order-model sets to better understand how IterativeImputer works. To view my notebook on Kaggle which includes the FULL descriptive tutorial, click here!

Be sure to follow me on GitHub and Kaggle where I'll be posting more notebooks and tutorials from time to time.
I've also created a Medium account to post tutorials and guides, please follow me there too.
If you think I deserve a coffee for the work I do, please do buy me a coffee!
Also, connect with me on LinkedIn.

Click here to check out my online portfolio!

As always, thanks again for stopping by.