Embedding training code in Power Query

One of the easiest solutions to train a machine learning model is to write the code needed to do so directly in Power Query, right after importing a dataset on which you will build the model.

Training a model on a fairly large dataset typically takes quite a bit of time to complete. As you embed the code in Power Query, it will run every time the data is refreshed, and this may result in a non-negligible delay in getting the data online. Hence, the following applies:

Important Note

This solution is recommended when you are certain that the time required to complete the model training is acceptable.

Let's now look at an example of how to write some training code using PyCaret.

Training and using ML models with PyCaret

Let's take the Titanic disaster dataset to train a machine learning model. Specifically, we want to create a model that predicts whether a passenger survives (the Survived column) based on their attributes...