proposal for model instances in future MCCs

View previous topic View next topic Go down

proposal for model instances in future MCCs

Post by crohr on Thu 27 Apr - 14:17

I would like to propose a modification on the number of instances for a model to be used in future MCCs.
It is clear that there is need for a reduction in the number of instances to be able to run the MCC.

I would suggest that for any model, having more than one instance, a maximum number of instances is provided in the following way.
1. a minimal instance
2. a maximum instance
3. a fixed number K (e.g. 2, 3 or 4) of instances equally distributed over the submitted model instances. There are models having more than one variable that is considered for the different instances and this should be reflected here too.
For example MAPK, ERK, Philosophers and many more have only one variable that is changed in the different instances, so for such models there should be K+2 instances.
For models with two variables to be considered in the scaling, e.g. PhaseVariation, DES, echo etc., there should be 2*K+2 instances and so on for models with more variables.
The number K should be the same for all models in the MCC, except for those models having already less than K instances.

This has the advantage that the bias of certain models having a large number of instances is reduced.

kind regards,
Christian

crohr

Posts : 3
Join date : 2017-02-07

View user profile

Back to top Go down

View previous topic View next topic Back to top

- Similar topics

 
Permissions in this forum:
You cannot reply to topics in this forum