hey Daniel,
So changing a transformation of a rigid body is done using the function set_transformation(), but it seems that the change actually happens only after model.evaluate() ( I am guessing after states update ?)
is there a way to force the changing on the transformation prior to model.evaluate() ?
thank you, Keren.
There is a (documented, I think) bool second argument which determines if it is done immediately or not. Is it not clear? Let me know if it can be made clearer. It might make more sense to have it as a separate function and have them refer to one another in their documentation).
On Sep 13, 2009, at 8:50 PM, Keren Lasker wrote:
> hey Daniel, > > So changing a transformation of a rigid body is done using the > function set_transformation(), but it seems that the change actually > happens only after model.evaluate() ( I am guessing after states > update ?) > > is there a way to force the changing on the transformation prior to > model.evaluate() ? > > > thank you, > Keren. > _______________________________________________ > IMP-dev mailing list > IMP-dev@salilab.org > https://salilab.org/mailman/listinfo/imp-dev
this is a bit confusing ... I think an explicit function can help avoiding bugs :) On Sep 22, 2009, at 4:21 PM, Daniel Russel wrote:
> There is a (documented, I think) bool second argument which > determines if it is done immediately or not. Is it not clear? Let me > know if it can be made clearer. It might make more sense to have it > as a separate function and have them refer to one another in their > documentation). > > On Sep 13, 2009, at 8:50 PM, Keren Lasker wrote: > >> hey Daniel, >> >> So changing a transformation of a rigid body is done using the >> function set_transformation(), but it seems that the change >> actually happens only after model.evaluate() ( I am guessing after >> states update ?) >> >> is there a way to force the changing on the transformation prior to >> model.evaluate() ? >> >> >> thank you, >> Keren. >> _______________________________________________ >> IMP-dev mailing list >> IMP-dev@salilab.org >> https://salilab.org/mailman/listinfo/imp-dev > > _______________________________________________ > IMP-dev mailing list > IMP-dev@salilab.org > https://salilab.org/mailman/listinfo/imp-dev
Since it keeps coming up, I would suggest: - set_transformation() which moves all the member particles - lazy_set_transformation() which does not.
Makes sense?
On Sep 22, 2009, at 5:23 PM, Keren Lasker wrote:
> this is a bit confusing ... > I think an explicit function can help avoiding bugs :) > On Sep 22, 2009, at 4:21 PM, Daniel Russel wrote: > >> There is a (documented, I think) bool second argument which >> determines if it is done immediately or not. Is it not clear? Let >> me know if it can be made clearer. It might make more sense to have >> it as a separate function and have them refer to one another in >> their documentation). >> >> On Sep 13, 2009, at 8:50 PM, Keren Lasker wrote: >> >>> hey Daniel, >>> >>> So changing a transformation of a rigid body is done using the >>> function set_transformation(), but it seems that the change >>> actually happens only after model.evaluate() ( I am guessing after >>> states update ?) >>> >>> is there a way to force the changing on the transformation prior >>> to model.evaluate() ? >>> >>> >>> thank you, >>> Keren. >>> _______________________________________________ >>> IMP-dev mailing list >>> IMP-dev@salilab.org >>> https://salilab.org/mailman/listinfo/imp-dev >> >> _______________________________________________ >> IMP-dev mailing list >> IMP-dev@salilab.org >> https://salilab.org/mailman/listinfo/imp-dev > > _______________________________________________ > IMP-dev mailing list > IMP-dev@salilab.org > https://salilab.org/mailman/listinfo/imp-dev
YES !! On Sep 22, 2009, at 6:09 PM, Daniel Russel wrote:
> Since it keeps coming up, I would suggest: > - set_transformation() which moves all the member particles > - lazy_set_transformation() which does not. > > Makes sense? > > On Sep 22, 2009, at 5:23 PM, Keren Lasker wrote: > >> this is a bit confusing ... >> I think an explicit function can help avoiding bugs :) >> On Sep 22, 2009, at 4:21 PM, Daniel Russel wrote: >> >>> There is a (documented, I think) bool second argument which >>> determines if it is done immediately or not. Is it not clear? Let >>> me know if it can be made clearer. It might make more sense to >>> have it as a separate function and have them refer to one another >>> in their documentation). >>> >>> On Sep 13, 2009, at 8:50 PM, Keren Lasker wrote: >>> >>>> hey Daniel, >>>> >>>> So changing a transformation of a rigid body is done using the >>>> function set_transformation(), but it seems that the change >>>> actually happens only after model.evaluate() ( I am guessing >>>> after states update ?) >>>> >>>> is there a way to force the changing on the transformation prior >>>> to model.evaluate() ? >>>> >>>> >>>> thank you, >>>> Keren. >>>> _______________________________________________ >>>> IMP-dev mailing list >>>> IMP-dev@salilab.org >>>> https://salilab.org/mailman/listinfo/imp-dev >>> >>> _______________________________________________ >>> IMP-dev mailing list >>> IMP-dev@salilab.org >>> https://salilab.org/mailman/listinfo/imp-dev >> >> _______________________________________________ >> IMP-dev mailing list >> IMP-dev@salilab.org >> https://salilab.org/mailman/listinfo/imp-dev > > _______________________________________________ > IMP-dev mailing list > IMP-dev@salilab.org > https://salilab.org/mailman/listinfo/imp-dev
Daniel - so it seems that even if set_transformation is forced the score states are updated, which means that safe_evaluate and evaluate will provide different scores - correct ? On Sep 22, 2009, at 6:13 PM, Keren Lasker wrote:
> YES !! > On Sep 22, 2009, at 6:09 PM, Daniel Russel wrote: > >> Since it keeps coming up, I would suggest: >> - set_transformation() which moves all the member particles >> - lazy_set_transformation() which does not. >> >> Makes sense? >> >> On Sep 22, 2009, at 5:23 PM, Keren Lasker wrote: >> >>> this is a bit confusing ... >>> I think an explicit function can help avoiding bugs :) >>> On Sep 22, 2009, at 4:21 PM, Daniel Russel wrote: >>> >>>> There is a (documented, I think) bool second argument which >>>> determines if it is done immediately or not. Is it not clear? Let >>>> me know if it can be made clearer. It might make more sense to >>>> have it as a separate function and have them refer to one another >>>> in their documentation). >>>> >>>> On Sep 13, 2009, at 8:50 PM, Keren Lasker wrote: >>>> >>>>> hey Daniel, >>>>> >>>>> So changing a transformation of a rigid body is done using the >>>>> function set_transformation(), but it seems that the change >>>>> actually happens only after model.evaluate() ( I am guessing >>>>> after states update ?) >>>>> >>>>> is there a way to force the changing on the transformation prior >>>>> to model.evaluate() ? >>>>> >>>>> >>>>> thank you, >>>>> Keren. >>>>> _______________________________________________ >>>>> IMP-dev mailing list >>>>> IMP-dev@salilab.org >>>>> https://salilab.org/mailman/listinfo/imp-dev >>>> >>>> _______________________________________________ >>>> IMP-dev mailing list >>>> IMP-dev@salilab.org >>>> https://salilab.org/mailman/listinfo/imp-dev >>> >>> _______________________________________________ >>> IMP-dev mailing list >>> IMP-dev@salilab.org >>> https://salilab.org/mailman/listinfo/imp-dev >> >> _______________________________________________ >> IMP-dev mailing list >> IMP-dev@salilab.org >> https://salilab.org/mailman/listinfo/imp-dev > > _______________________________________________ > IMP-dev mailing list > IMP-dev@salilab.org > https://salilab.org/mailman/listinfo/imp-dev
On Sep 22, 2009, at 10:07 PM, Keren Lasker wrote:
> Daniel - > so it seems that even if set_transformation is forced the score > states are updated, which means that safe_evaluate and evaluate will > provide different scores - correct ? safe_evaluate will disappear very shortly as evaluate itself will be made safe. Once the changes are made to set_transformation then rigid body members will have the updated coordinates immediately and, at least that part of, the update score states in Model::evaluate will be a no-op.
> On Sep 22, 2009, at 6:13 PM, Keren Lasker wrote: > >> YES !! >> On Sep 22, 2009, at 6:09 PM, Daniel Russel wrote: >> >>> Since it keeps coming up, I would suggest: >>> - set_transformation() which moves all the member particles >>> - lazy_set_transformation() which does not. >>> >>> Makes sense? >>> >>> On Sep 22, 2009, at 5:23 PM, Keren Lasker wrote: >>> >>>> this is a bit confusing ... >>>> I think an explicit function can help avoiding bugs :) >>>> On Sep 22, 2009, at 4:21 PM, Daniel Russel wrote: >>>> >>>>> There is a (documented, I think) bool second argument which >>>>> determines if it is done immediately or not. Is it not clear? >>>>> Let me know if it can be made clearer. It might make more sense >>>>> to have it as a separate function and have them refer to one >>>>> another in their documentation). >>>>> >>>>> On Sep 13, 2009, at 8:50 PM, Keren Lasker wrote: >>>>> >>>>>> hey Daniel, >>>>>> >>>>>> So changing a transformation of a rigid body is done using the >>>>>> function set_transformation(), but it seems that the change >>>>>> actually happens only after model.evaluate() ( I am guessing >>>>>> after states update ?) >>>>>> >>>>>> is there a way to force the changing on the transformation >>>>>> prior to model.evaluate() ? >>>>>> >>>>>> >>>>>> thank you, >>>>>> Keren. >>>>>> _______________________________________________ >>>>>> IMP-dev mailing list >>>>>> IMP-dev@salilab.org >>>>>> https://salilab.org/mailman/listinfo/imp-dev >>>>> >>>>> _______________________________________________ >>>>> IMP-dev mailing list >>>>> IMP-dev@salilab.org >>>>> https://salilab.org/mailman/listinfo/imp-dev >>>> >>>> _______________________________________________ >>>> IMP-dev mailing list >>>> IMP-dev@salilab.org >>>> https://salilab.org/mailman/listinfo/imp-dev >>> >>> _______________________________________________ >>> IMP-dev mailing list >>> IMP-dev@salilab.org >>> https://salilab.org/mailman/listinfo/imp-dev >> >> _______________________________________________ >> IMP-dev mailing list >> IMP-dev@salilab.org >> https://salilab.org/mailman/listinfo/imp-dev > > _______________________________________________ > IMP-dev mailing list > IMP-dev@salilab.org > https://salilab.org/mailman/listinfo/imp-dev
-
Daniel Russel -
Keren Lasker