Number of feature space splits = max_depth +1?

Hi, maybe I misunderstood but I was under the impression that the number of feature space splits is basically max_depth + 1. So a max_depth of 1 should give two splits, as we see in the first tree regression figure in this notebook.

Nevertheless, the attached figure was generated using a max_depth=3, yet the feature space is splited in more than 4 times.

Thanks
image

The hyperparameter max_depth and some other are better explained in the notebook Importance of decision tree hyperparameter later in this module. There we show for instance the following decision tree:

It has max_depth=2 but 4 splits.