rcrpy.NonParametric

class rcrpy.NonParametric

Bases: object

Base class for user-supplied non-parametric mu computation.

Subclass and override mu_func (and/or mu_func_w) to control which points contribute to the central-value estimate at each iteration of the RCR rejection loop.

The indices attribute records the post-mu_func index list; the rejection loop reads it back after each call.

__init__()

Methods

__init__()
mu_func(flags, y)	Return (indices, true_y) given current flags and the raw y array.
mu_func_w(flags, w, y)	Weighted twin of mu_func.

mu_func(flags, y)

Return (indices, true_y) given current flags and the raw y array.

Default behavior mirrors the C++ no-op base case: keep all flagged points. Subclasses override for non-trivial filtering.

Parameters:

• flags (ndarray)

• y (ndarray)

Return type:

tuple[ndarray, ndarray]

mu_func_w(flags, w, y)

Weighted twin of mu_func. Returns (indices, true_w, true_y).

Parameters:

• flags (ndarray)

• w (ndarray)

• y (ndarray)

Return type:

tuple[ndarray, ndarray, ndarray]