parallel runs

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

parallel runs

cking
In Weka, I am confused as to what it means by the parallelRuns option in
Auto-WEKA. Can you help me?


I know it says 'the number of runs to perform in parallel' in  the notes but
does this mean it divides up the dataset and processes it in parallel? Or
what exactly?



--
Sent from: https://weka.8497.n7.nabble.com/
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to [hidden email]
To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit https://list.waikato.ac.nz/postorius/lists/wekalist.list.waikato.ac.nz
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|

Re: parallel runs

Peter Reutemann-3
On March 14, 2020 1:02:53 AM GMT+13:00, cking <[hidden email]> wrote:

>In Weka, I am confused as to what it means by the parallelRuns option
>in
>Auto-WEKA. Can you help me?
>
>
>I know it says 'the number of runs to perform in parallel' in  the
>notes but
>does this mean it divides up the dataset and processes it in parallel?
>Or
>what exactly?
>
>
>
>--
>Sent from: https://weka.8497.n7.nabble.com/
>_______________________________________________
>Wekalist mailing list -- [hidden email]
>Send posts to [hidden email]
>To unsubscribe send an email to [hidden email]
>To subscribe, unsubscribe, etc., visit
>https://list.waikato.ac.nz/postorius/lists/wekalist.list.waikato.ac.nz
>List etiquette:
>http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

Unlikely. I'd say it will evaluate different configurations/setups in parallel to speed up the process.

Cheers, Peter
--
Peter Reutemann
Dept. of Computer Science
University of Waikato, NZ
+64 (7) 858-5174
http://www.cms.waikato.ac.nz/~fracpete/
http://www.data-mining.co.nz/
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to [hidden email]
To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit https://list.waikato.ac.nz/postorius/lists/wekalist.list.waikato.ac.nz
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|

Re: parallel runs

Michael Hall
In reply to this post by cking


On Mar 13, 2020, at 7:02 AM, cking <[hidden email]> wrote:

In Weka, I am confused as to what it means by the parallelRuns option in
Auto-WEKA. Can you help me?


I know it says 'the number of runs to perform in parallel' in  the notes but
does this mean it divides up the dataset and processes it in parallel? Or
what exactly?

From…

Possibly this…

Finally, we note that both TPE and SMAC are randomized algorithms and thus expected to produce different results based on the random seed provided. As demonstrated in [16], this allows for trivial, yet effective parallelization of the op- timization process: simply perform k independent runs of the optimization method in parallel and select the result of the run with the lowest cross-validation error to return.3 We verified experimentally that the more parallel runs, the faster this process identifies high-quality configurations; neverthe- less, we restricted the version of Auto-WEKA studied here to run only 4 parallel processes, in order to study a setting typical for commonly used workstations.

Appears to be an experimental feature. 

_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to [hidden email]
To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit https://list.waikato.ac.nz/postorius/lists/wekalist.list.waikato.ac.nz
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html