Description

When reading the API documentation of CalibratedClassifierCV I had a hard time understanding how it works and how to properly use it without actually looking at the code.

The main reasons were:

1. that the suffix CV made me think (as in other cross-validation estimators) cross-validation is used to either evaluate a metric or tune hyperparameters, while
   in this case it is used to create multiple calibrated classifiers whose predictions are then averaged for prediction.
2. the lack of examples in the API doc demonstrating the two completely different use cases, depending on cv='prefit', which actually made the CV suffix even more confusing for me.
   See also this question on stackoverflow.

I think documentation could be changed to make it easier to understand.