Experimenter

I am running Experimenter with "Distribute" on multiple RemoteEngines.
I want to know if it is possible to use a number of RemoteEngines which is
larger than the total number of Runs.

For instance, suppose I do 10-fold cross validation and "From 1 To 10" runs
per data set  and algorithm.

Suppose I have 10 algorithms and 10 data sets.

So if I understand, in total, there will be 10 runs (each of 10-fold cross
validation) for each of the 10x10=100 combinations (where a combination is
(algorithm, dataset)).

So 1,000 runs of 10-fold cross validation per run.

Now, suppose I have 25 cores (and I launched 25 RemoteEngines on the
computer).

From my experiments, it appears that only 10 remote engines are in use for
executing all the runs. It appears that the number of RemoteEngines in use
= the number of runs per combination. In the above, 10 runs per combination.

Is this correct?

It seems that it would be more beneficial if I could use all my cores, say,
25 cores, to run all the 1,000 runs, rather than use just 10 cores.

Is it possible to configure this?


thanks
Joel







-- 
Prof. Joel Ratsaby
http://www.ariel.ac.il/sites/ratsaby
Electrical and Electronics Engineering Dept.
Ariel University
ARIEL 40700
ISRAEL

_______________________________________________
Wekalist mailing list -- [email protected]
Send posts to [email protected]
To unsubscribe send an email to [email protected]
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