Hi all, Is there a way for me to find a profile using a projection as a data source? In particular, I'd like to find a radial profile of a surface density in a disk galaxy simulation. I think I could do this by histogramming a fixed resolution buffer created by a projection data source but is that the only way to do it? I'd prefer for my data pipeline to look identical for computing profiles and in all other cases I'm using a Profile1D object to find things like the cumulative mass or the average density as a function of radius. Thanks for your help with this! -Nathan
Hi Nathan, On Wed, Apr 30, 2014 at 8:35 PM, Nathan Goldbaum <nathan12343@gmail.com> wrote:
Hi all,
Is there a way for me to find a profile using a projection as a data source? In particular, I'd like to find a radial profile of a surface density in a disk galaxy simulation.
I think I could do this by histogramming a fixed resolution buffer created by a projection data source but is that the only way to do it? I'd prefer for my data pipeline to look identical for computing profiles and in all other cases I'm using a Profile1D object to find things like the cumulative mass or the average density as a function of radius.
I believe this is what has been done in the past -- Sam wrote a set of scripts that did it this way. I think, though, that you might be able to use the cylindrical radius derived fields and then accumulate within them. Would that get to where you want the analysis? -Matt
Thanks for your help with this!
-Nathan
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
I have some scripts for doing projected radius profiles using projections and FRBs. They'll certainly be different from the Profile1D stuff, but they work OK. I'm happy to share them if you want a starter for this avenue. Cameron On Wed, Apr 30, 2014 at 6:24 PM, Matthew Turk <matthewturk@gmail.com> wrote:
Hi Nathan,
Hi all,
Is there a way for me to find a profile using a projection as a data
In particular, I'd like to find a radial profile of a surface density in a disk galaxy simulation.
I think I could do this by histogramming a fixed resolution buffer created by a projection data source but is that the only way to do it? I'd
On Wed, Apr 30, 2014 at 8:35 PM, Nathan Goldbaum <nathan12343@gmail.com> wrote: source? prefer
for my data pipeline to look identical for computing profiles and in all other cases I'm using a Profile1D object to find things like the cumulative mass or the average density as a function of radius.
I believe this is what has been done in the past -- Sam wrote a set of scripts that did it this way. I think, though, that you might be able to use the cylindrical radius derived fields and then accumulate within them. Would that get to where you want the analysis?
-Matt
Thanks for your help with this!
-Nathan
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
-- Cameron Hummels Postdoctoral Researcher Steward Observatory University of Arizona http://chummels.org
On Thu, May 1, 2014 at 8:07 AM, Cameron Hummels <chummels@gmail.com> wrote:
I have some scripts for doing projected radius profiles using projections and FRBs. They'll certainly be different from the Profile1D stuff, but they work OK. I'm happy to share them if you want a starter for this avenue.
Would love to see them.
Cameron
On Wed, Apr 30, 2014 at 6:24 PM, Matthew Turk <matthewturk@gmail.com>wrote:
Hi Nathan,
Hi all,
Is there a way for me to find a profile using a projection as a data
In particular, I'd like to find a radial profile of a surface density in a disk galaxy simulation.
I think I could do this by histogramming a fixed resolution buffer created by a projection data source but is that the only way to do it? I'd
On Wed, Apr 30, 2014 at 8:35 PM, Nathan Goldbaum <nathan12343@gmail.com> wrote: source? prefer
for my data pipeline to look identical for computing profiles and in all other cases I'm using a Profile1D object to find things like the cumulative mass or the average density as a function of radius.
I believe this is what has been done in the past -- Sam wrote a set of scripts that did it this way. I think, though, that you might be able to use the cylindrical radius derived fields and then accumulate within them. Would that get to where you want the analysis?
-Matt
Thanks for your help with this!
-Nathan
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
-- Cameron Hummels Postdoctoral Researcher Steward Observatory University of Arizona http://chummels.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
It would be cool if we were able to treat a projection as a 2D data container where you could define a center, create a radius field, and then just pipe directly into the regular profiling machinery. How hard would it be to do that (longer timescale than meeting Nathan's needs here of course)? On Thu, May 1, 2014 at 2:34 PM, Nathan Goldbaum <nathan12343@gmail.com>wrote:
On Thu, May 1, 2014 at 8:07 AM, Cameron Hummels <chummels@gmail.com>wrote:
I have some scripts for doing projected radius profiles using projections and FRBs. They'll certainly be different from the Profile1D stuff, but they work OK. I'm happy to share them if you want a starter for this avenue.
Would love to see them.
Cameron
On Wed, Apr 30, 2014 at 6:24 PM, Matthew Turk <matthewturk@gmail.com>wrote:
Hi Nathan,
Hi all,
Is there a way for me to find a profile using a projection as a data
In particular, I'd like to find a radial profile of a surface density in a disk galaxy simulation.
I think I could do this by histogramming a fixed resolution buffer created by a projection data source but is that the only way to do it? I'd
On Wed, Apr 30, 2014 at 8:35 PM, Nathan Goldbaum <nathan12343@gmail.com> wrote: source? prefer
for my data pipeline to look identical for computing profiles and in all other cases I'm using a Profile1D object to find things like the cumulative mass or the average density as a function of radius.
I believe this is what has been done in the past -- Sam wrote a set of scripts that did it this way. I think, though, that you might be able to use the cylindrical radius derived fields and then accumulate within them. Would that get to where you want the analysis?
-Matt
Thanks for your help with this!
-Nathan
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
-- Cameron Hummels Postdoctoral Researcher Steward Observatory University of Arizona http://chummels.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
Hi Britton, On Sun, May 4, 2014 at 9:25 AM, Britton Smith <brittonsmith@gmail.com> wrote:
It would be cool if we were able to treat a projection as a 2D data container where you could define a center, create a radius field, and then just pipe directly into the regular profiling machinery. How hard would it be to do that (longer timescale than meeting Nathan's needs here of course)?
I agree. So I think the only reason this would be difficult is because of the degeneracy between asking for a derived field that would be derived *in the grids* and asking for a derived field that would be derived *in the data object*. Typically, we always assume that the data object can generate derived fields -- effectively, what this means is that if you ask for velocity magnitude, the velocity magnitude can be derived from the x-y-z velocities in the data object and that it doesn't need to go back to the individual chunks/grids/particles. This is how slices work, but projections don't because we assume the act of projecting is how the data is generated. ("Construction" versus "selection" data objects.) So if we can break that degeneracy -- having a way to say, this object *has* everything we need to do the derived fields, so do them in it, then I think we can implement it. I think it's mostly a nomenclature thing. What if we were able to make projections serve as a source to another data object? Would that suffice?
On Thu, May 1, 2014 at 2:34 PM, Nathan Goldbaum <nathan12343@gmail.com> wrote:
On Thu, May 1, 2014 at 8:07 AM, Cameron Hummels <chummels@gmail.com> wrote:
I have some scripts for doing projected radius profiles using projections and FRBs. They'll certainly be different from the Profile1D stuff, but they work OK. I'm happy to share them if you want a starter for this avenue.
Would love to see them.
Cameron
On Wed, Apr 30, 2014 at 6:24 PM, Matthew Turk <matthewturk@gmail.com> wrote:
Hi Nathan,
On Wed, Apr 30, 2014 at 8:35 PM, Nathan Goldbaum <nathan12343@gmail.com> wrote:
Hi all,
Is there a way for me to find a profile using a projection as a data source? In particular, I'd like to find a radial profile of a surface density in a disk galaxy simulation.
I think I could do this by histogramming a fixed resolution buffer created by a projection data source but is that the only way to do it? I'd prefer for my data pipeline to look identical for computing profiles and in all other cases I'm using a Profile1D object to find things like the cumulative mass or the average density as a function of radius.
I believe this is what has been done in the past -- Sam wrote a set of scripts that did it this way. I think, though, that you might be able to use the cylindrical radius derived fields and then accumulate within them. Would that get to where you want the analysis?
-Matt
Thanks for your help with this!
-Nathan
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
-- Cameron Hummels Postdoctoral Researcher Steward Observatory University of Arizona http://chummels.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
Hi Matt, Ok, that makes a lot of sense. I think your idea would work. If we made a projection become a 2D dataset with only the fields that were projected available as the base fields, this could be quite straightforward. Britton On Sun, May 4, 2014 at 1:34 PM, Matthew Turk <matthewturk@gmail.com> wrote:
Hi Britton,
It would be cool if we were able to treat a projection as a 2D data container where you could define a center, create a radius field, and
On Sun, May 4, 2014 at 9:25 AM, Britton Smith <brittonsmith@gmail.com> wrote: then
just pipe directly into the regular profiling machinery. How hard would it be to do that (longer timescale than meeting Nathan's needs here of course)?
I agree. So I think the only reason this would be difficult is because of the degeneracy between asking for a derived field that would be derived *in the grids* and asking for a derived field that would be derived *in the data object*. Typically, we always assume that the data object can generate derived fields -- effectively, what this means is that if you ask for velocity magnitude, the velocity magnitude can be derived from the x-y-z velocities in the data object and that it doesn't need to go back to the individual chunks/grids/particles. This is how slices work, but projections don't because we assume the act of projecting is how the data is generated. ("Construction" versus "selection" data objects.)
So if we can break that degeneracy -- having a way to say, this object *has* everything we need to do the derived fields, so do them in it, then I think we can implement it. I think it's mostly a nomenclature thing. What if we were able to make projections serve as a source to another data object? Would that suffice?
On Thu, May 1, 2014 at 2:34 PM, Nathan Goldbaum <nathan12343@gmail.com> wrote:
On Thu, May 1, 2014 at 8:07 AM, Cameron Hummels <chummels@gmail.com> wrote:
I have some scripts for doing projected radius profiles using
projections
and FRBs. They'll certainly be different from the Profile1D stuff, but they work OK. I'm happy to share them if you want a starter for this avenue.
Would love to see them.
Cameron
On Wed, Apr 30, 2014 at 6:24 PM, Matthew Turk <matthewturk@gmail.com> wrote:
Hi Nathan,
On Wed, Apr 30, 2014 at 8:35 PM, Nathan Goldbaum <
nathan12343@gmail.com>
wrote:
Hi all,
Is there a way for me to find a profile using a projection as a data source? In particular, I'd like to find a radial profile of a surface density in a disk galaxy simulation.
I think I could do this by histogramming a fixed resolution buffer created by a projection data source but is that the only way to do it? I'd prefer for my data pipeline to look identical for computing profiles and in all other cases I'm using a Profile1D object to find things like the cumulative mass or the average density as a function of radius.
I believe this is what has been done in the past -- Sam wrote a set of scripts that did it this way. I think, though, that you might be able to use the cylindrical radius derived fields and then accumulate within them. Would that get to where you want the analysis?
-Matt
Thanks for your help with this!
-Nathan
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
-- Cameron Hummels Postdoctoral Researcher Steward Observatory University of Arizona http://chummels.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
Hi all, Coming in a little late to the game on this, but I wanted to point out that this workflow would work (currently off the tip of my yt-3.x repo): http://nbviewer.ipython.org/gist/jzuhone/11539419 the only issue being that you have to use AstroPy and FITS as a way-station. Best, John On May 5, 2014, at 7:44 AM, Britton Smith <brittonsmith@gmail.com> wrote:
Hi Matt,
Ok, that makes a lot of sense. I think your idea would work. If we made a projection become a 2D dataset with only the fields that were projected available as the base fields, this could be quite straightforward.
Britton
On Sun, May 4, 2014 at 1:34 PM, Matthew Turk <matthewturk@gmail.com> wrote: Hi Britton,
On Sun, May 4, 2014 at 9:25 AM, Britton Smith <brittonsmith@gmail.com> wrote:
It would be cool if we were able to treat a projection as a 2D data container where you could define a center, create a radius field, and then just pipe directly into the regular profiling machinery. How hard would it be to do that (longer timescale than meeting Nathan's needs here of course)?
I agree. So I think the only reason this would be difficult is because of the degeneracy between asking for a derived field that would be derived *in the grids* and asking for a derived field that would be derived *in the data object*. Typically, we always assume that the data object can generate derived fields -- effectively, what this means is that if you ask for velocity magnitude, the velocity magnitude can be derived from the x-y-z velocities in the data object and that it doesn't need to go back to the individual chunks/grids/particles. This is how slices work, but projections don't because we assume the act of projecting is how the data is generated. ("Construction" versus "selection" data objects.)
So if we can break that degeneracy -- having a way to say, this object *has* everything we need to do the derived fields, so do them in it, then I think we can implement it. I think it's mostly a nomenclature thing. What if we were able to make projections serve as a source to another data object? Would that suffice?
On Thu, May 1, 2014 at 2:34 PM, Nathan Goldbaum <nathan12343@gmail.com> wrote:
On Thu, May 1, 2014 at 8:07 AM, Cameron Hummels <chummels@gmail.com> wrote:
I have some scripts for doing projected radius profiles using projections and FRBs. They'll certainly be different from the Profile1D stuff, but they work OK. I'm happy to share them if you want a starter for this avenue.
Would love to see them.
Cameron
On Wed, Apr 30, 2014 at 6:24 PM, Matthew Turk <matthewturk@gmail.com> wrote:
Hi Nathan,
On Wed, Apr 30, 2014 at 8:35 PM, Nathan Goldbaum <nathan12343@gmail.com> wrote:
Hi all,
Is there a way for me to find a profile using a projection as a data source? In particular, I'd like to find a radial profile of a surface density in a disk galaxy simulation.
I think I could do this by histogramming a fixed resolution buffer created by a projection data source but is that the only way to do it? I'd prefer for my data pipeline to look identical for computing profiles and in all other cases I'm using a Profile1D object to find things like the cumulative mass or the average density as a function of radius.
I believe this is what has been done in the past -- Sam wrote a set of scripts that did it this way. I think, though, that you might be able to use the cylindrical radius derived fields and then accumulate within them. Would that get to where you want the analysis?
-Matt
Thanks for your help with this!
-Nathan
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
-- Cameron Hummels Postdoctoral Researcher Steward Observatory University of Arizona http://chummels.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
participants (5)
-
Britton Smith -
Cameron Hummels -
John ZuHone -
Matthew Turk -
Nathan Goldbaum