Data rarely comes in usable form. Data wrangling and exploratory data analysis are the difference ... This is easy to implement with standard Python libraries. Which imputation strategy is best?