I’ve been wanting to try out the competitions from DrivenData and thought this might be a relatively straightforward challenge to try. Check them out for other environmentally and socially conscious data-driven competitions
Also wanted to see what works for making a data science cookiecutter template featuring streamlit for interactive data exploration, reports, prediction, and even training in some cases!
- Prediction on test dataset
- Pandas profiling with the excellent Pandas Profiling
- Exploratory Analysis of features and labels
- Data source information
- Feature engineering application steps
- Catalog of model training