Error: Probability is invalid (-0.11111111111111116) in state (R1=7,.......... (line 101, column 33).

[Ramesh Kumar]

I am receiving the above error for couple of experiments eventhough I just standard IPD transition dilemma.

Is this a serious error or Prism takes care of such error through approximations.

Thanks for support,      ...Ramesh

[Ramesh Kumar]

This happens only when w=0.90 or above while setting the number of rounds as below:

const int rounds=round(1/(1-w)); 

The above feeds into another formula which decides one of the transition probabilities as below:

formula sizeBAD =  l/((rounds-1)+l-g)

[Dave Parker]

I guess it also depends on the value of g, Ramesh?

Send the model if you are still stuck.

Best wishes,

Dave

> --
> You received this message because you are subscribed to the Google
> Groups "PRISM model checker" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to prismmodelchec...@googlegroups.com
> <mailto:prismmodelchec...@googlegroups.com>.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/prismmodelchecker/e46ea32c-e9c6-4b2d-90af-d55aa5fcdadfn%40googlegroups.com
> <https://groups.google.com/d/msgid/prismmodelchecker/e46ea32c-e9c6-4b2d-90af-d55aa5fcdadfn%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Ramesh Kumar]

Thanks,David. I got this cleared. Appreciate your support during my learning period.

Ramesh