24093511909
91%
main: 92%

Ran 07 Apr 2026 05:03PM UTC

Jobs 1

Files 49

Run time 1min

Badge

Embed ▾

Committed 07 Apr 2026 04:51PM UTC coverage: 91.231%. Remained the same

Build # 24093511909

Build Type

push

github

Committed by f-dangel

Commit Message

[REF] Always measure compiled performance in benchmark

Remove `_IS_COMPILABLE` gate — always measure compiled matvec,
compiled precompute phases, and compiled peak memory for all
operators. Those without compile support will simply show no speedup.

Deduplicate compiled precompute timing: instead of a separate
`_time_compiled_kfac_factors` method, wrap each precompute phase
in `torch.compile` and re-time the full pipeline.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Coverage Stats

2424 of 2657 relevant lines covered (91.23%)

0.91 hits per line

Jobs

ID	Job ID	Ran	Files	Coverage
1	24093511909.1	07 Apr 2026 05:03PM UTC	49	91.23	GitHub Action Run

f-dangel / curvlinops / 24093511909
91%
main: 92%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Jobs

Source Files on build 24093511909

f-dangel / curvlinops / 24093511909 91% main: 92%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Jobs

Source Files on build 24093511909

f-dangel / curvlinops / 24093511909
91%
main: 92%

README BADGES
x