halo finding on gadget snapshots
Hi All,
I'm having some issues with really basic halo finding on gadget snapshots in the bleeding edge version. What I want to do is ID halos in a volume, save a list of the halos, and plot a projection plot with circles over the halos (all of which I think should be possible).
The principle script is: http://paste.yt-project.org/show/4824/
The error I get when trying to save the halos is:
http://paste.yt-project.org/show/4822/
and in trying to annotate the halos on the projection plot is:
http://paste.yt-project.org/show/4823/
Am I either doing something wrong in my commands? (hopefully as simple as that...) or are these not supported yet in the bleeding edge yt?
thanks desika
ps. On another (related) note, is it possible to manually draw circles on a projection plot? Or arrows, or tick marks, or anything that I could use to visually show off individual halos?
On Tue, Jun 24, 2014 at 8:57 AM, Desika Narayanan dnarayan@haverford.edu wrote:
Hi All,
I'm having some issues with really basic halo finding on gadget snapshots in the bleeding edge version. What I want to do is ID halos in a volume, save a list of the halos, and plot a projection plot with circles over the halos (all of which I think should be possible).
The principle script is: http://paste.yt-project.org/show/4824/
The error I get when trying to save the halos is:
I don't think I can speak to this beyond that this error indicates that something is going wrong with field detection. It might be helpful to rerun the script with yt's log level threshold decreased to 10, which will show debug messages. See this page: http://yt-project.org/docs/dev-3.0/reference/configuration.html
Running with debug logs will allow us to see if any fields are raising exceptions during field detection.
and in trying to annotate the halos on the projection plot is:
http://paste.yt-project.org/show/4823/
Am I either doing something wrong in my commands? (hopefully as simple as that...) or are these not supported yet in the bleeding edge yt?
This looks like a straight up typo in the HaloCatalogCallback. You could try commenting out the line it's failing on.
It looks like Hillary Egan was the last one to touch this in May.
thanks desika
ps. On another (related) note, is it possible to manually draw circles on a projection plot? Or arrows, or tick marks, or anything that I could use to visually show off individual halos?
You can access the underlying matplotlib primitives like so:
slc = SlicePlot(pf, 0, 'density') plot = slc.plots['density'] fig = plot.figure # matplotlib figure object ax = plot.axes # matplotlib axes object cax = plot.cax # matplotlib colorbar axes object cb = plot.cb # matplotlib colorbar object image = plot.image # matplotlib AxesImage object
Using these objects you should be able to customize the plot however you'd like.
-Nathan
yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
Hi Desika,
This is almost certainly a problem with the old halo finders (anything that you import from analysis_modules.halo_finding) not being completely compatible with yt-3.0 at the moment. The best way to access their functionality now is to go through the new HaloCatalog api. I have only gone through the old halo catalog methods to fix them enough to work with the new HaloCatalog api, and didn't test them in general with the idea that they will be phased out as stand alone tools.
Here is an example script for finding the halos using a halo catalog: http://paste.yt-project.org/show/4826/
If you'd like more information about how the halo catalog works, its mostly documented here: http://yt-project.org/docs/dev-3.0/analyzing/analysis_modules/halo_catalogs....
Hopefully that works for you! -Hilary
On Tue, Jun 24, 2014 at 11:39 AM, Nathan Goldbaum nathan12343@gmail.com wrote:
On Tue, Jun 24, 2014 at 8:57 AM, Desika Narayanan dnarayan@haverford.edu wrote:
Hi All,
I'm having some issues with really basic halo finding on gadget snapshots in the bleeding edge version. What I want to do is ID halos in a volume, save a list of the halos, and plot a projection plot with circles over the halos (all of which I think should be possible).
The principle script is: http://paste.yt-project.org/show/4824/
The error I get when trying to save the halos is:
I don't think I can speak to this beyond that this error indicates that something is going wrong with field detection. It might be helpful to rerun the script with yt's log level threshold decreased to 10, which will show debug messages. See this page: http://yt-project.org/docs/dev-3.0/reference/configuration.html
Running with debug logs will allow us to see if any fields are raising exceptions during field detection.
and in trying to annotate the halos on the projection plot is:
http://paste.yt-project.org/show/4823/
Am I either doing something wrong in my commands? (hopefully as simple as that...) or are these not supported yet in the bleeding edge yt?
This looks like a straight up typo in the HaloCatalogCallback. You could try commenting out the line it's failing on.
It looks like Hillary Egan was the last one to touch this in May.
thanks desika
ps. On another (related) note, is it possible to manually draw circles on a projection plot? Or arrows, or tick marks, or anything that I could use to visually show off individual halos?
You can access the underlying matplotlib primitives like so:
slc = SlicePlot(pf, 0, 'density') plot = slc.plots['density'] fig = plot.figure # matplotlib figure object ax = plot.axes # matplotlib axes object cax = plot.cax # matplotlib colorbar axes object cb = plot.cb # matplotlib colorbar object image = plot.image # matplotlib AxesImage object
Using these objects you should be able to customize the plot however you'd like.
-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
Hi Hilary and Nathan,
Thanks a lot for the info on the new 3.0 halo modules.
I tried running your script, Hilary, and got the following errors: http://paste.yt-project.org/show/4831/
Any thoughts/help would be greatly appreciated! I can re-run with increased log level like Nathan mentioned if that's useful.
thanks, desika
On Tue, Jun 24, 2014 at 2:05 PM, Hilary Egan hilaryye@gmail.com wrote:
Hi Desika,
This is almost certainly a problem with the old halo finders (anything that you import from analysis_modules.halo_finding) not being completely compatible with yt-3.0 at the moment. The best way to access their functionality now is to go through the new HaloCatalog api. I have only gone through the old halo catalog methods to fix them enough to work with the new HaloCatalog api, and didn't test them in general with the idea that they will be phased out as stand alone tools.
Here is an example script for finding the halos using a halo catalog: http://paste.yt-project.org/show/4826/
If you'd like more information about how the halo catalog works, its mostly documented here: http://yt-project.org/docs/dev-3.0/analyzing/analysis_modules/halo_catalogs....
Hopefully that works for you! -Hilary
On Tue, Jun 24, 2014 at 11:39 AM, Nathan Goldbaum nathan12343@gmail.com wrote:
On Tue, Jun 24, 2014 at 8:57 AM, Desika Narayanan <dnarayan@haverford.edu
wrote:
Hi All,
I'm having some issues with really basic halo finding on gadget snapshots in the bleeding edge version. What I want to do is ID halos in a volume, save a list of the halos, and plot a projection plot with circles over the halos (all of which I think should be possible).
The principle script is: http://paste.yt-project.org/show/4824/
The error I get when trying to save the halos is:
I don't think I can speak to this beyond that this error indicates that something is going wrong with field detection. It might be helpful to rerun the script with yt's log level threshold decreased to 10, which will show debug messages. See this page: http://yt-project.org/docs/dev-3.0/reference/configuration.html
Running with debug logs will allow us to see if any fields are raising exceptions during field detection.
and in trying to annotate the halos on the projection plot is:
http://paste.yt-project.org/show/4823/
Am I either doing something wrong in my commands? (hopefully as simple as that...) or are these not supported yet in the bleeding edge yt?
This looks like a straight up typo in the HaloCatalogCallback. You could try commenting out the line it's failing on.
It looks like Hillary Egan was the last one to touch this in May.
thanks desika
ps. On another (related) note, is it possible to manually draw circles on a projection plot? Or arrows, or tick marks, or anything that I could use to visually show off individual halos?
You can access the underlying matplotlib primitives like so:
slc = SlicePlot(pf, 0, 'density') plot = slc.plots['density'] fig = plot.figure # matplotlib figure object ax = plot.axes # matplotlib axes object cax = plot.cax # matplotlib colorbar axes object cb = plot.cb # matplotlib colorbar object image = plot.image # matplotlib AxesImage object
Using these objects you should be able to customize the plot however you'd like.
-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
yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
participants (3)
-
Desika Narayanan -
Hilary Egan -
Nathan Goldbaum