Impossibility to use 'from last model'

questions concerning analysis/theory using programs M-SURGE, E-SURGE and U-CARE

Impossibility to use 'from last model'

Postby Deborah » Wed Aug 31, 2011 10:41 am

Hi Rémi,

I am getting troubles with local minima and wanted to try the option 'from last model' for initial values since some models manage to converge, instead of running each model with 10 multiple random values.
However after running the model that works (with the output registered in Excel file as well as .out), when I change the inits to 'from last model' the IVFV tells me that it was not possible and that the inits will be kept constant...
any idea why I cannot use this function?
I know that some other persons managed to use it, maybe Iam making something wrong in the order at which things should be done...

Thanks a lot
Deborah
Deborah
 
Posts: 3
Joined: Tue Mar 30, 2010 1:44 am
Location: Chizé, France

Re: Impossibility to use 'from last model'

Postby CHOQUET » Mon Sep 05, 2011 5:01 am

Hello Deborah,
Normally to use the option 'from last model', you have to follow the following steps.
1) Retrieve a model that you want to use for IV
2) Select the option 'from last model'
3) Go to gemaco, build the new model
4) Go to ivfv and select the [0,1] scale to fix the IV and run
Rémi
CHOQUET
 
Posts: 211
Joined: Thu Nov 24, 2005 4:58 am
Location: CEFE, Montpellier, FRANCE.

Re: Impossibility to use 'from last model'

Postby Deborah » Wed Sep 07, 2011 3:32 am

Hi Rémi,
I managed to run using "from last model", but by running the model from which I want to take the inits right before (which takes a lot of time), it would be better to use the retrieve function but it gets stuck after calling the gemaco,
I get this error message: "Subscript indices must either be real positive integers or logicals."
DO you know where the problem could be?
Thanks a lot
Deborah
Deborah
 
Posts: 3
Joined: Tue Mar 30, 2010 1:44 am
Location: Chizé, France

Re: Impossibility to use 'from last model'

Postby CHOQUET » Fri Sep 09, 2011 8:25 am

Hello,

If there is several fixed parameters, there is a bug while saving the Excel file.
Perhaps, it is the main reason for the retrieve model failure.
But it is possible to modify the Excel file by hand, while waiting for the new version of E-SURGE.

Open the Excel file, go to the IVFV thumb. If there is a mistake it should be like this (more than two rows in the biological axis, in the example 4)

Fix(values) : biological axis 1
Fix(values) : biological axis 0
Fix(indices) : biological axis 2
Fix(indices) : biological axis 4

Modify the ivfv thumb according to this (so that only two rows appears for biological axis, bringing the two first rows and the two last rows together)

Fix(values) : biological axis 1 0
Fix(indices) : biological axis 2 4

Sincerely,
Rémi
CHOQUET
 
Posts: 211
Joined: Thu Nov 24, 2005 4:58 am
Location: CEFE, Montpellier, FRANCE.


Return to analysis help

Who is online

Users browsing this forum: No registered users and 19 guests