Dear Alps developers,
I would like to ask you a simple question about the simulation output of the alps scheduler:
Is there a straightforward way to extract the average value of a given observable as a function of the monte carlo time? (for example, to roughly see wether the equilibration phase is completed). I couldn't find an example of this on the online tutorial.
Thank you in advance,
Giuseppe
On 31 Aug 2010, at 23:30, Giuseppe Carleo wrote:
Dear Alps developers,
I would like to ask you a simple question about the simulation output of the alps scheduler:
Is there a straightforward way to extract the average value of a given observable as a function of the monte carlo time? (for example, to roughly see wether the equilibration phase is completed). I couldn't find an example of this on the online tutorial.
Thank you in advance,
Giuseppe
Dear Giuseppe,
the information is there on a coarse time scale and one could make that finer but we do not plot it yet. We do have plans to add an easy way to make such plots in the future. Are you interested and want to program this?
Matthias
Dear Giuseppe,
the information is there on a coarse time scale and one could make that finer but we do not plot it yet. We do have plans to add an easy way to make such plots in the future. Are you interested and want to program this?
Matthias
Dear Prof. Troyer,
I'd be glad to implement it. I think that there should be an option for the scheduler in order to enable (only if needed) a regular tracking of the averages. I also think that, as a first attempt, it would be enough to have recorded only the averages computed at the checkpoints... This should be easy to implement, but if I find problems we can probably discuss about this in a few days, in Zurich.
Best,
Giuseppe
On 4 Sep 2010, at 19:24, Giuseppe Carleo wrote:
Dear Giuseppe,
the information is there on a coarse time scale and one could make that finer but we do not plot it yet. We do have plans to add an easy way to make such plots in the future. Are you interested and want to program this?
Matthias
Dear Prof. Troyer,
I'd be glad to implement it. I think that there should be an option for the scheduler in order to enable (only if needed) a regular tracking of the averages. I also think that, as a first attempt, it would be enough to have recorded only the averages computed at the checkpoints... This should be easy to implement, but if I find problems we can probably discuss about this in a few days, in Zurich.
Actually we store the averages of 32-64 bins and you will be able to make plots of those averages. Let us discuss this in Zurich.
Matthias
comp-phys-alps-users@lists.phys.ethz.ch