Learner1D's bound check algo in self.ask doesn't take self.data or self.pending_points about adaptive

[basnijholt]

originally posted by Jorn Hoofwijk (@Jorn) at 2018-09-15T12:45:31.109Z on GitLab

I was thinking, maybe we could add the bound points, and if you requested more points, then do a call to ask again.

in pseudocode:

def ask(n):
    bounds_todo = <something>
    if bounds_todo:
        self.tell_many(bounds_todo, None)
        extra_points = self.ask(n - len(bounds_todo))
        return bounds_todo + extra_points
    else:
        return <the non-bound points that are interesting>

[basnijholt]

originally posted by Jorn Hoofwijk (@Jorn) at 2018-09-15T12:53:27.986Z on GitLab

wait a sec, this only works if add_data=True

[basnijholt]

originally posted by Bas Nijholt (@basnijholt) at 2018-09-15T13:09:31.283Z on GitLab

wait a sec, this only works if add_data=True

Exactly.