This repository includes the codes for the my talk at PyData Eindhoven 2020: The industrial challenge of missing data. The talk is available on YouTube, and the slides are here.
Two practical notebooks ready to run in colab are available to practice missing data imputation.
The notebook "PyData_missingdata.ipynb" includes some imputation examples as shown and explained in the talk. It may get extended with new methodologies, stay tuned!
The notebook "PyData_GAIN.ipynb" is the implementation of GAIN with kersa and Tensorflow 2.x. Noting that another different implementation of this is available here
[Sklearn.impute] (https://scikit-learn.org/stable/modules/impute.html)
[fancyimpute] (https://github.com/iskandr/fancyimpute)