Profile ARIMA to identify performance bottlenecks
Offen
performance
spike
- Vorherrschende Sprache
- Python
- Sterne
- 850
- Forks
- 96
- PR-Merge-Kennzahlen
- Keine gemergten PRs in 30 T.
Beschreibung
In #2365, we noticed that some arima tests took about a minute to run despite only running on 500 rows.
@ParthivNaresh thinks it may be because the starting set of parameters for ARIMA are not great.
This issue tracks profiling ARIMA to see where the bottleneck is and proposing fixes.
Original comment from PR:
> I'm surprised this test took ~60 seconds on the original data size since its 500 rows.
>
> I think we should file a separate issue to see where the bottleneck is.
>
> _Originally posted by @freddyaboulton in https://github.com/alteryx/evalml/pull/2365#discussion_r651955382_
Beitragsleitfaden
Bewertung
Dieses Issue wurde noch nicht bewertet.