Clean up #272
Clean up #272
Conversation
mohamed-barakat
force-pushed
the
CategoryOfColimitQuivers
branch
from
74b11bc
to
79c92ca
Compare
mohamed-barakat
force-pushed
the
CategoryOfColimitQuivers
branch
from
79c92ca
to
eb42997
Compare
| @@ -43,7 +43,7 @@ gap> runtime_quiver := Runtime( ) - start;; | |||
| # | |||
| gap> if runtime >= runtime_quiver * 3 / 10 then Display( true ); else Display( runtime ); Display( runtime_quiver ); fi; | |||
| true | |||
| gap> if runtime <= runtime_quiver * 4 / 10 then Display( true ); else Display( runtime ); Display( runtime_quiver ); fi; | |||
| gap> if runtime <= runtime_quiver * 5 / 10 then Display( true ); else Display( runtime ); Display( runtime_quiver ); fi; | |||
There was a problem hiding this comment.
If this is needed, this means that the compiled code has gotten slower compared to the non-compiled code. Which commit is responsible for this?
There was a problem hiding this comment.
I can only reproduce this on my laptop, where the noncompiled code seems to be faster than in the CI.
There was a problem hiding this comment.
Has this always been the case, i.e. already when this test was introduced?
There was a problem hiding this comment.
I will wait for your answer before I merge.
There was a problem hiding this comment.
Awkward enough, the regression is caused by PR homalg-project/HigherHomologicalAlgebra#158:
Being on #271
- The regerssion does not occur with PR Complexes as Presheaves HigherHomologicalAlgebra#156
- The regerssion occurs PR added locales to the list of needed packages for ComplexesCategories … HigherHomologicalAlgebra#158
There was a problem hiding this comment.
Can you post runtime and runtime_quiver before and after the regression?
There was a problem hiding this comment.
Without regression:
AdditiveClosureOfAlgebroid_vs_QuiverRows
msecs: 3367
gap> runtime;
647
gap> runtime_quiver;
1647
gap> Float( runtime_quiver * 4 / 10 );
658.8With regression:
AdditiveClosureOfAlgebroid_vs_QuiverRows
msecs: 3316
gap> runtime;
666
gap> runtime_quiver;
1586
gap> Float( runtime_quiver * 4 / 10 );
634.4There was a problem hiding this comment.
Sorry, my first numbers were wrong, now they are fixed.
There was a problem hiding this comment.
You see from the timings that with PR homalg-project/HigherHomologicalAlgebra#158 the runtime of QuiverRows is dropping.
Codecov ReportPatch coverage:
Additional details and impacted files@@ Coverage Diff @@
## master #272 +/- ##
==========================================
- Coverage 73.50% 73.50% -0.01%
==========================================
Files 322 322
Lines 53174 53170 -4
==========================================
- Hits 39085 39081 -4
Misses 14089 14089
Flags with carried forward coverage won't be shown. Click here to find out more.
☔ View full report in Codecov by Sentry. |
No description provided.