Introduce LASSO benchmarks

[ClaudiaComito]

Due Diligence

• General:
  • title of the PR is suitable to appear in the Release Notes
• Implementation:
  • unit tests: all split configurations tested
  • unit tests: multiple dtypes tested
  • NEW unit tests: MPS tested (1 MPI process, 1 GPU)
  • benchmarks: created for new functionality
  • benchmarks: performance improved or maintained
  • documentation updated where needed

Description

Issue/s resolved: #2175

Changes proposed:

• introduced regression.py (with many thanks to Gemini 3)
• updated main.py accordingly

Type of change

Memory requirements

Performance

Does this change modify the behaviour of other functions? If so, which?

yes / no

[ClaudiaComito]

Tagging @mrfh92 who introduced the performance improvements in #2056 . Does the LASSO benchmark make sense? Not sure about the split=1 use case. Removed it as not supported. Thanks Fabian!

[JuanPedroGHM]

Benchmarks results - Sponsored by perun

FTPx86-AMD-Milan-Mi250 - CPU - N4 - RUNTIME

Function	Value [s]	Change %	30 day avg. [s]	Last entrie [s]
lanczos	0.519957	11.3408	0.466996	0.467504
randomized_svd_split1	0.0174002	-26.5893	0.0237026	0.0228314
randomized_eigh_split0	0.0166024	26.0445	0.0131718	0.0115408
kmedoids	0.361845	10.5865	0.327206	0.327008
reshape	0.226448	19.5351	0.189441	0.195912
apply_inplace_min_max_scaler_and_inverse	0.00294886	-34.1959	0.00448126	0.00325899
apply_inplace_max_abs_scaler_and_inverse	0.000582504	-48.9229	0.00114044	0.00249567
apply_inplace_normalizer	0.0076231	16.8565	0.00652347	0.00350266

---

FTPx86-AMD-Milan-Mi250 - GPU - N4 - RUNTIME

Function	Value [s]	Change %	30 day avg. [s]	Last entrie [s]
randomized_svd_split1	0.078686	12.4214	0.069992	0.0677005
apply_inplace_min_max_scaler_and_inverse	0.00252161	-48.4336	0.00489003	0.0061677
apply_inplace_max_abs_scaler_and_inverse	0.00261707	37.7912	0.0018993	0.00227404
apply_inplace_normalizer	0.0095871	-11.9776	0.0108917	0.0111076
convolution_array_distributed	0.0197586	14.8941	0.0171972	0.015065

---

Grafana Dashboard
Last updated: 2026-03-13T05:12:30Z

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 91.63%. Comparing base (2ca98d5) to head (baf34ca).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2176   +/-   ##
=======================================
  Coverage   91.63%   91.63%           
=======================================
  Files          86       86           
  Lines       14005    14007    +2     
=======================================
+ Hits        12833    12835    +2     
  Misses       1172     1172           

Flag	Coverage Δ
unit	91.63% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.