Using ColumnTransformer to combine data processing steps
Create cohesive pipelines for processing data where different columns require different techniques
Published in
6 min readFeb 22, 2020
This scikit-learn tool comes in extremely handy, but also has some quirks of its own. Today we’ll be using it to transform data on ferry wait time for the Edmonds-Kingston route of the Washington State Ferries. (Thank you WSF for the data!). Full disclosure: we’re just going to use a small portion of the data set today.