estimator.util.local_minimum

class estimator.util.local_minimum(start, stop, precision=1, smallerf=<function local_minimum.<lambda>>, suppress_bounds_warning=False, log_level=5)

An iterator context for finding a local minimum using binary search.

We use the neighborhood of a point to decide the next direction to go into (gradient descent style), so the algorithm is not plain binary search (see update() function.)

We also zoom out by a factor precision, find an approximate local minimum and then search the neighbourhood for the smallest value.

Note

We combine an iterator and a context to give the caller access to the result.

Create a fresh local minimum search context.

Parameters:

• start – starting point

• stop – end point (exclusive)

• precision – only consider every precision-th value in the main loop

• smallerf – a function to decide if lhs is smaller than rhs.

• suppress_bounds_warning – do not warn if a boundary is picked as optimal

__call__(**kwargs)

Call self as a function.

Methods

__init__(start, stop[, precision, smallerf, ...])	Create a fresh local minimum search context.
update(res)	TESTS:

Attributes

neighborhood	An iterator over the neighborhood of the currently best value.
x
y