Re: [math] discuss faster getEntry for RealMatrix implementations
Le sam. 4 mars 2023 à 17:08, Cyril de Catheu
a écrit :
> [...]
> > > >
> > > Let me know what you think.
> >
> > I've also added comments on JIRA.
> >
>
> Somehow I can't see any comments on JIRA -
> https://issues.apache.org/jira/browse/MATH-1654
> Am I missing something?
I wrote them... And forgot to click on "add". :-{
Done now.
Regards,
Gilles
-
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org
Re: [math] The build is broken
Is it consistently broken? One test for the optimisers requires 15 restarts due to randomness in the optimiser and the difficulty of the optimisation problem. This runs in a different phase. See the surefire plug-in configuration in the math parent pom. We tried to make the test suite more robust but you still get failures. Gilles did a lot of work on the simplex tests which by their nature are random in their search process. The rest of the test suite runs with up to 5 retries and sporadically fails. But most of the time what we have will pass so I did not move more tests to the difficulty to pass group. If you check the build log you can see if the build failed on one of these optimisation tests or if it is failing somewhere else. Alex On Sat, 4 Mar 2023, 15:21 sebb, wrote: > On Sat, 4 Mar 2023 at 15:16, Gary Gregory wrote: > > > > Let me get this straight: You use Gmail but you refuse to sign in to > > GitHub? Right... > > So what? > > They are run by different organisations. > The consequences of a leak are very different, especially if you only > use GMail for public mails. > > Let people choose what 3rd party services they want to use. > > > Gary > > > > On Sat, Mar 4, 2023, 10:11 Gilles Sadowski wrote: > > > > > Hello Gary. > > > > > > Le sam. 4 mars 2023 à 15:52, Gary D. Gregory a > > > écrit : > > > > > > > > Hi Math, > > > > > > > > The build is broken locally for me, and as exemplified on GHA [1]: > > > > https://github.com/apache/commons-math/actions/runs/4086106809/jobs/7044975362 > > > > > > Thanks for the notice but a button says "Sign in to view logs". > > > No-go. > > > > > > > > > > > Please fix. > > > > > > Could you please post the logs in a JIRA report? > > > > > > Two builds in a row were successful here: > > > > > > $ git pull > > > Already up to date. > > > $ mvn > > > [...] > > > [INFO] Reactor Summary for Apache Commons Math 4.0-SNAPSHOT: > > > [INFO] > > > [INFO] Apache Commons Math SUCCESS [ > > > 4.842 s] > > > [INFO] Miscellaneous core classes . SUCCESS [ > > > 15.431 s] > > > [INFO] Artificial neural networks . SUCCESS [ > > > 9.253 s] > > > [INFO] Transforms . SUCCESS [ > > > 7.217 s] > > > [INFO] Exception classes (Legacy) . SUCCESS [ > > > 7.430 s] > > > [INFO] Miscellaneous core classes (Legacy) SUCCESS [ > > > 16.820 s] > > > [INFO] Apache Commons Math (Legacy) ... SUCCESS > [01:24 > > > min] > > > [INFO] Apache Commons Math Documentation .. SUCCESS [ > > > 5.144 s] > > > [INFO] Example applications ... SUCCESS [ > > > 0.212 s] > > > [INFO] SOFM ... SUCCESS [ > > > 0.274 s] > > > [INFO] SOFM: Chinese Rings SUCCESS [ > > > 5.772 s] > > > [INFO] SOFM: Traveling Salesman Problem ... SUCCESS [ > > > 5.743 s] > > > [INFO] K-Means SUCCESS [ > > > 0.487 s] > > > [INFO] K-Means: Image Clustering .. SUCCESS [ > > > 5.322 s] > > > [INFO] > > > > > > > [INFO] BUILD SUCCESS > > > [INFO] > > > > > > > [INFO] Total time: 02:48 min > > > [INFO] Finished at: 2023-03-04T16:05:58+01:00 > > > [INFO] > > > > > > > > > > Regards, > > > Gilles > > > > > > > > > > > Gary > > > > [1] https://github.com/apache/commons-math/actions > > > > > > > > > > - > > > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org > > > For additional commands, e-mail: dev-h...@commons.apache.org > > > > > > > > - > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org > For additional commands, e-mail: dev-h...@commons.apache.org > >
Re: [math] discuss faster getEntry for RealMatrix implementations
Thanks for the quick answer! ↓ On Sat, Mar 4, 2023 at 2:39 PM Gilles Sadowski wrote: > Hello. > > Le sam. 4 mars 2023 à 12:29, Cyril de Catheu > a écrit : > > > > Hello, > > > > I just opened a ticket > > Thanks for the report! > > > describing an opportunity to optimize the getEntry > > method of RealMatrix implementations. > > For our specific workload in BOBYQA > > Advertisement: Help is very welcome for maintaining the > implementation of that algorithm.[1] > Ok I had a look at the tickets, looks interesting. I'd be happy to help maintain the implementation. I've recently worked on java code generated from fortran. It seems to be the case in this implementation - special mention for the infinite loop + switch acting as a GOTO x) . I'll start by refactoring. > > this seems to be a ~10% speed > > improvement. > > I think it could also give a (variable) speed up to every consumer of > > RealMatrix.getEntry. > > > > https://issues.apache.org/jira/browse/MATH-1654 > > Thanks for the analysis. > > Side note: You are most welcome to contribute to having JMH > benchmarks integrated within the component's repository.[2] > Noted. Maybe later. > > If changing RealMatrix implems is not an option, there are other specific > > solutions to optimize the BOBYQA optimizer. > > > > Let me know what you think. > > I've also added comments on JIRA. > Somehow I can't see any comments on JIRA - https://issues.apache.org/jira/browse/MATH-1654 Am I missing something? Regards. Cyril > Best regards, > Gilles > > [1] Cf. related issues in JIRA (and in the ML archive). > [2] By coincidence, another discussion has just been restarted about > it (for another functionality): > https://issues.apache.org/jira/browse/MATH-1585 > > - > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org > For additional commands, e-mail: dev-h...@commons.apache.org > >
Re: [math] The build is broken
On Sat, 4 Mar 2023 at 15:16, Gary Gregory wrote: > > Let me get this straight: You use Gmail but you refuse to sign in to > GitHub? Right... So what? They are run by different organisations. The consequences of a leak are very different, especially if you only use GMail for public mails. Let people choose what 3rd party services they want to use. > Gary > > On Sat, Mar 4, 2023, 10:11 Gilles Sadowski wrote: > > > Hello Gary. > > > > Le sam. 4 mars 2023 à 15:52, Gary D. Gregory a > > écrit : > > > > > > Hi Math, > > > > > > The build is broken locally for me, and as exemplified on GHA [1]: > > https://github.com/apache/commons-math/actions/runs/4086106809/jobs/7044975362 > > > > Thanks for the notice but a button says "Sign in to view logs". > > No-go. > > > > > > > > Please fix. > > > > Could you please post the logs in a JIRA report? > > > > Two builds in a row were successful here: > > > > $ git pull > > Already up to date. > > $ mvn > > [...] > > [INFO] Reactor Summary for Apache Commons Math 4.0-SNAPSHOT: > > [INFO] > > [INFO] Apache Commons Math SUCCESS [ > > 4.842 s] > > [INFO] Miscellaneous core classes . SUCCESS [ > > 15.431 s] > > [INFO] Artificial neural networks . SUCCESS [ > > 9.253 s] > > [INFO] Transforms . SUCCESS [ > > 7.217 s] > > [INFO] Exception classes (Legacy) . SUCCESS [ > > 7.430 s] > > [INFO] Miscellaneous core classes (Legacy) SUCCESS [ > > 16.820 s] > > [INFO] Apache Commons Math (Legacy) ... SUCCESS [01:24 > > min] > > [INFO] Apache Commons Math Documentation .. SUCCESS [ > > 5.144 s] > > [INFO] Example applications ... SUCCESS [ > > 0.212 s] > > [INFO] SOFM ... SUCCESS [ > > 0.274 s] > > [INFO] SOFM: Chinese Rings SUCCESS [ > > 5.772 s] > > [INFO] SOFM: Traveling Salesman Problem ... SUCCESS [ > > 5.743 s] > > [INFO] K-Means SUCCESS [ > > 0.487 s] > > [INFO] K-Means: Image Clustering .. SUCCESS [ > > 5.322 s] > > [INFO] > > > > [INFO] BUILD SUCCESS > > [INFO] > > > > [INFO] Total time: 02:48 min > > [INFO] Finished at: 2023-03-04T16:05:58+01:00 > > [INFO] > > > > > > Regards, > > Gilles > > > > > > > > Gary > > > [1] https://github.com/apache/commons-math/actions > > > > > > > - > > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org > > For additional commands, e-mail: dev-h...@commons.apache.org > > > > - To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
Re: [math] The build is broken
Let me get this straight: You use Gmail but you refuse to sign in to GitHub? Right... Gary On Sat, Mar 4, 2023, 10:11 Gilles Sadowski wrote: > Hello Gary. > > Le sam. 4 mars 2023 à 15:52, Gary D. Gregory a > écrit : > > > > Hi Math, > > > > The build is broken locally for me, and as exemplified on GHA [1]: > https://github.com/apache/commons-math/actions/runs/4086106809/jobs/7044975362 > > Thanks for the notice but a button says "Sign in to view logs". > No-go. > > > > > Please fix. > > Could you please post the logs in a JIRA report? > > Two builds in a row were successful here: > > $ git pull > Already up to date. > $ mvn > [...] > [INFO] Reactor Summary for Apache Commons Math 4.0-SNAPSHOT: > [INFO] > [INFO] Apache Commons Math SUCCESS [ > 4.842 s] > [INFO] Miscellaneous core classes . SUCCESS [ > 15.431 s] > [INFO] Artificial neural networks . SUCCESS [ > 9.253 s] > [INFO] Transforms . SUCCESS [ > 7.217 s] > [INFO] Exception classes (Legacy) . SUCCESS [ > 7.430 s] > [INFO] Miscellaneous core classes (Legacy) SUCCESS [ > 16.820 s] > [INFO] Apache Commons Math (Legacy) ... SUCCESS [01:24 > min] > [INFO] Apache Commons Math Documentation .. SUCCESS [ > 5.144 s] > [INFO] Example applications ... SUCCESS [ > 0.212 s] > [INFO] SOFM ... SUCCESS [ > 0.274 s] > [INFO] SOFM: Chinese Rings SUCCESS [ > 5.772 s] > [INFO] SOFM: Traveling Salesman Problem ... SUCCESS [ > 5.743 s] > [INFO] K-Means SUCCESS [ > 0.487 s] > [INFO] K-Means: Image Clustering .. SUCCESS [ > 5.322 s] > [INFO] > > [INFO] BUILD SUCCESS > [INFO] > > [INFO] Total time: 02:48 min > [INFO] Finished at: 2023-03-04T16:05:58+01:00 > [INFO] > > > Regards, > Gilles > > > > > Gary > > [1] https://github.com/apache/commons-math/actions > > > > - > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org > For additional commands, e-mail: dev-h...@commons.apache.org > >
Re: [math] The build is broken
Hello Gary. Le sam. 4 mars 2023 à 15:52, Gary D. Gregory a écrit : > > Hi Math, > > The build is broken locally for me, and as exemplified on GHA [1]: > https://github.com/apache/commons-math/actions/runs/4086106809/jobs/7044975362 Thanks for the notice but a button says "Sign in to view logs". No-go. > > Please fix. Could you please post the logs in a JIRA report? Two builds in a row were successful here: $ git pull Already up to date. $ mvn [...] [INFO] Reactor Summary for Apache Commons Math 4.0-SNAPSHOT: [INFO] [INFO] Apache Commons Math SUCCESS [ 4.842 s] [INFO] Miscellaneous core classes . SUCCESS [ 15.431 s] [INFO] Artificial neural networks . SUCCESS [ 9.253 s] [INFO] Transforms . SUCCESS [ 7.217 s] [INFO] Exception classes (Legacy) . SUCCESS [ 7.430 s] [INFO] Miscellaneous core classes (Legacy) SUCCESS [ 16.820 s] [INFO] Apache Commons Math (Legacy) ... SUCCESS [01:24 min] [INFO] Apache Commons Math Documentation .. SUCCESS [ 5.144 s] [INFO] Example applications ... SUCCESS [ 0.212 s] [INFO] SOFM ... SUCCESS [ 0.274 s] [INFO] SOFM: Chinese Rings SUCCESS [ 5.772 s] [INFO] SOFM: Traveling Salesman Problem ... SUCCESS [ 5.743 s] [INFO] K-Means SUCCESS [ 0.487 s] [INFO] K-Means: Image Clustering .. SUCCESS [ 5.322 s] [INFO] [INFO] BUILD SUCCESS [INFO] [INFO] Total time: 02:48 min [INFO] Finished at: 2023-03-04T16:05:58+01:00 [INFO] Regards, Gilles > > Gary > [1] https://github.com/apache/commons-math/actions > - To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
[math] The build is broken
Hi Math, The build is broken locally for me, and as exemplified on GHA [1]: https://github.com/apache/commons-math/actions/runs/4086106809/jobs/7044975362 Please fix. Gary [1] https://github.com/apache/commons-math/actions - To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
Re: [math] discuss faster getEntry for RealMatrix implementations
Hello. Le sam. 4 mars 2023 à 12:29, Cyril de Catheu a écrit : > > Hello, > > I just opened a ticket Thanks for the report! > describing an opportunity to optimize the getEntry > method of RealMatrix implementations. > For our specific workload in BOBYQA Advertisement: Help is very welcome for maintaining the implementation of that algorithm.[1] > this seems to be a ~10% speed > improvement. > I think it could also give a (variable) speed up to every consumer of > RealMatrix.getEntry. > > https://issues.apache.org/jira/browse/MATH-1654 Thanks for the analysis. Side note: You are most welcome to contribute to having JMH benchmarks integrated within the component's repository.[2] > If changing RealMatrix implems is not an option, there are other specific > solutions to optimize the BOBYQA optimizer. > > Let me know what you think. I've also added comments on JIRA. Best regards, Gilles [1] Cf. related issues in JIRA (and in the ML archive). [2] By coincidence, another discussion has just been restarted about it (for another functionality): https://issues.apache.org/jira/browse/MATH-1585 - To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
[math] discuss faster getEntry for RealMatrix implementations
Hello, I just opened a ticket describing an opportunity to optimize the getEntry method of RealMatrix implementations. For our specific workload in BOBYQA this seems to be a ~10% speed improvement. I think it could also give a (variable) speed up to every consumer of RealMatrix.getEntry. https://issues.apache.org/jira/browse/MATH-1654 If changing RealMatrix implems is not an option, there are other specific solutions to optimize the BOBYQA optimizer. Let me know what you think. Have a nice week-end. Cyril de Catheu
