Having performed this in turicreate in jupyter noteboopk:
small_model = turicreate.decision_tree_classifier.create(train_data,
validation_set=None,
target = target,
features = features,
max_depth = 2)
I am unclear what the visualization route is.
I use this package which uses other modules.
But if I try sklearn, then it looks like the whole process must be in sklearn, as an example. That is not what I want. Looking for an add on that accepts this output as input. But may be that is not possible.
Not possible. You need to write a python routine to print the decision tree in an awkward manner.