hanfei1986/Impute-missing-data-with-KNNImputer-and-IterativeImputer
When signaficant amount of data are missing, what can we do? Impute the missing data with mean or median? Actually, Scikit-Learn provides two powerful imputers, KNNImputer and IterativeImputer, which can do this work effectively.
Impute-missing-data-with-KNNImputer-and-IterativeImputer
When signaficant amount of data are missing, what can we do? Impute the missing data with mean or median? That will be a diaster. Actually, Scikit-Learn provides two powerful imputers, KNNImputer and IterativeImputer. The former imputes missing data using the mean value from n_neighbors nearest neighbors found in the training set, and the latter is inspired by R's MICE package and imputes missing values by modeling each feature with missing values as a function of other features in a round-robin fashion.
Before imputation, there are significant amount of "Cost", a few "Weight", and many "Ingredient Number" data missing in the dataset.
After imputation, all the columns are filled.
Let's have a look at the imputation effect. Amazing!