This concludes the section on feature engineering.
You now have some hands-on experience scrubbing, scaling, transforming, binning, one-hot encoding and crossing feature data with MLNET and C#. You’ll use these skills in later labs to optimize your machine learning predictions.
The exact sequence of data transformation steps has a huge impact on the accuracy of machine learning predictions. This is why feature engineering is such an important step in data science.
As you practice with more and more datasets, you will slowly build an intuition for choosing the right transformation step for each feature column in your data.
But until then, just remember to always normalize your data and one-hot encode anything that looks like a category!