Hello,
I have a dataset containing both categorical and continuous variables. Both types of variables contain missing data.
I would like to impute the missing data, but I'm not sure how.
I am thinking of doing the following:

1. first - impute categorical data with a simple imputer (with 'constant' or 'most_frequent')
2. After inputting the categorical data, perform one-hot encoding for categorical variables
3. Then, impute continuous data with KNN.

Do you think this is a good idea? Any other suggestions?