Chapter 3: cross_val_predict()

This chapter uses `cross_val_predict()` to make cross validation predictions, and then use them to compute some metrics like confusion matrix, recall, precision, and more.

But Sklearn documentation (https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.cross_val_predict.html) says the following: "It is not appropriate to pass these predictions into an evaluation metric. Use cross_validate to measure generalization error."

So, what's the point here? Can we use `cross_val_predict()` results to compute metrics? Or am I missing something?

Thanks in advance!