Suggesting a minor enhancement in mpdist.py:: _mpdist_vect

I think we can replace

https://github.com/TDAmeritrade/stumpy/blob/423c679fdfcdd547744d2b9a2aed7565124fd6e1/stumpy/mpdist.py#L100-L106

with

P_ABBA = np.empty(2 * j, dtype=np.float64) 
  
 if k is None: 
     percentage = np.clip(percentage, 0.0, 1.0) 
     k = math.ceil(percentage * (2 * Q.shape[0]))    # BECAUSE `P_ABBA.shape[0] == 2 * j`, see the first line 
  
 k = min(int(k), P_ABBA.shape[0] - 1)

Gotcha. I think that is reasonable. We'll need to adjust aampdist as well.

Opened #1158 with this, including the aampdist counterpart you mentioned. Since P_ABBA = np.empty(2 * j), the following k = min(int(k), P_ABBA.shape[0] - 1) already applies the same 2 * j - 1 bound, so the inner min is redundant. Verified it's behaviour-preserving over 20,000 randomised (n, m, percentage) combinations (including out-of-range percentages so the clip path is hit) — identical k every time — and test_mpdist.py/test_aampdist.py pass (82 tests).

Fixed stumpy-dev#1082 Drop redundant k clamp in _mpdist_vect and _aam…

Navigation Menu

Repository navigation

Suggesting a minor enhancement in `mpdist.py:: _mpdist_vect` #1082

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Suggesting a minor enhancement in mpdist.py:: _mpdist_vect #1082

Description

Activity

seanlaw commented on Apr 20, 2025

DMZ22 commented on Jul 24, 2026

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions

Suggesting a minor enhancement in `mpdist.py:: _mpdist_vect` #1082