I'm trying to plot 3D streamlines from a FLASH (v.4) dataset. I've based my streamlines script off the version presented on the yt-project.org website.
I consistently get an error about going up the ghost cells in my volume. The error message is:
RuntimeError: Error: yt attempted to read outside the boundaries of a
non-periodic domain along dimension 0. Region left edge = -3e+18 code_length, Region right edge = 4.5e+18 code_length Dataset left edge = 0.0 code_length, Dataset right edge = 1.2e+19 code_length This commonly happens when trying to compute ghost cells up to the domain boundary. Two possible solutions are to load a smaller region that does not border the edge or override the periodicity for this dataset.
My FLASH datasets have x/y/zmin = 0.0 and x/y/zmax 1.2e19 cm.
I seems like yt is trying to read outside my domain, or is bumping up against the guardcells and I have no idea how to prevent it from doing this. I have tried inspecting the code for an obvious solution. A very similar error shows up when I try to make 1D profiles of the entire dataset (dd = ds.all_data()).
is it possible that yt doesn't support volumes where the (0,0,0) coordinate is a corner of the volume? Any easy solution to the above issue?
Thanks, Mikhail
Hi Mikhail,
Sorry you're having trouble--any chance you can post your script using paste.yt-project.org or a similar paste service? It'll be easier to figure out what's going on then.
By the way, what are your simulation's boundary conditions?
Best,
John Z
John ZuHone Kavli Center for Astrophysics and Space Research Massachusetts Institute of Technology 77 Massachusetts Ave., 37-582G Cambridge, MA 02139 (w) 617-253-2354 (m) 781-708-5004 jzuhone@space.mit.edu jzuhone@gmail.com http://www.jzuhone.com
On Jul 30, 2015, at 9:30 PM, Mikhail Klassen klassm@mcmaster.ca wrote:
I'm trying to plot 3D streamlines from a FLASH (v.4) dataset. I've based my streamlines script off the version presented on the yt-project.org website.
I consistently get an error about going up the ghost cells in my volume. The error message is:
RuntimeError: Error: yt attempted to read outside the boundaries of a non-periodic domain along dimension 0. Region left edge = -3e+18 code_length, Region right edge = 4.5e+18 code_length Dataset left edge = 0.0 code_length, Dataset right edge = 1.2e+19 code_length This commonly happens when trying to compute ghost cells up to the domain boundary. Two possible solutions are to load a smaller region that does not border the edge or override the periodicity for this dataset.
My FLASH datasets have x/y/zmin = 0.0 and x/y/zmax 1.2e19 cm.
I seems like yt is trying to read outside my domain, or is bumping up against the guardcells and I have no idea how to prevent it from doing this. I have tried inspecting the code for an obvious solution. A very similar error shows up when I try to make 1D profiles of the entire dataset (dd = ds.all_data()).
is it possible that yt doesn't support volumes where the (0,0,0) coordinate is a corner of the volume? Any easy solution to the above issue?
Thanks, Mikhail _______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
You can do:
ds.periodicity = (1, 1, 1)
Technically this is wrong for your dataset, but if you don't care what happens at the boundary it should be fine.
There's an open issue to do boundary conditions for yt's ghost zones properly, but so far no one has taken it up.
Nathan
On Thursday, July 30, 2015, John ZuHone jzuhone@gmail.com wrote:
Hi Mikhail,
Sorry you're having trouble--any chance you can post your script using paste.yt-project.org or a similar paste service? It'll be easier to figure out what's going on then.
By the way, what are your simulation's boundary conditions?
Best,
John Z
John ZuHone Kavli Center for Astrophysics and Space Research Massachusetts Institute of Technology 77 Massachusetts Ave., 37-582G Cambridge, MA 02139 (w) 617-253-2354 (m) 781-708-5004 jzuhone@space.mit.edu javascript:_e(%7B%7D,'cvml','jzuhone@space.mit.edu'); jzuhone@gmail.com javascript:_e(%7B%7D,'cvml','jzuhone@gmail.com'); http://www.jzuhone.com
On Jul 30, 2015, at 9:30 PM, Mikhail Klassen <klassm@mcmaster.ca javascript:_e(%7B%7D,'cvml','klassm@mcmaster.ca');> wrote:
I'm trying to plot 3D streamlines from a FLASH (v.4) dataset. I've based my streamlines script off the version presented on the yt-project.org website.
I consistently get an error about going up the ghost cells in my volume. The error message is:
RuntimeError: Error: yt attempted to read outside the boundaries of a
non-periodic domain along dimension 0. Region left edge = -3e+18 code_length, Region right edge = 4.5e+18 code_length Dataset left edge = 0.0 code_length, Dataset right edge = 1.2e+19 code_length This commonly happens when trying to compute ghost cells up to the domain boundary. Two possible solutions are to load a smaller region that does not border the edge or override the periodicity for this dataset.
My FLASH datasets have x/y/zmin = 0.0 and x/y/zmax 1.2e19 cm.
I seems like yt is trying to read outside my domain, or is bumping up against the guardcells and I have no idea how to prevent it from doing this. I have tried inspecting the code for an obvious solution. A very similar error shows up when I try to make 1D profiles of the entire dataset (dd = ds.all_data()).
is it possible that yt doesn't support volumes where the (0,0,0) coordinate is a corner of the volume? Any easy solution to the above issue?
Thanks, Mikhail
yt-users mailing list yt-users@lists.spacepope.org javascript:_e(%7B%7D,'cvml','yt-users@lists.spacepope.org'); http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
Argh, sorry, Mike is right, it should be (True, True, True).
Does the error message not show the explicit code example? If not it probably should, "override the periodicity" isn't clear enough unless you know more about yt internals than we should expect.
On Thursday, July 30, 2015, Nathan Goldbaum nathan12343@gmail.com wrote:
You can do:
ds.periodicity = (1, 1, 1)
Technically this is wrong for your dataset, but if you don't care what happens at the boundary it should be fine.
There's an open issue to do boundary conditions for yt's ghost zones properly, but so far no one has taken it up.
Nathan
On Thursday, July 30, 2015, John ZuHone <jzuhone@gmail.com javascript:_e(%7B%7D,'cvml','jzuhone@gmail.com');> wrote:
Hi Mikhail,
Sorry you're having trouble--any chance you can post your script using paste.yt-project.org or a similar paste service? It'll be easier to figure out what's going on then.
By the way, what are your simulation's boundary conditions?
Best,
John Z
John ZuHone Kavli Center for Astrophysics and Space Research Massachusetts Institute of Technology 77 Massachusetts Ave., 37-582G Cambridge, MA 02139 (w) 617-253-2354 (m) 781-708-5004 jzuhone@space.mit.edu jzuhone@gmail.com http://www.jzuhone.com
On Jul 30, 2015, at 9:30 PM, Mikhail Klassen klassm@mcmaster.ca wrote:
I'm trying to plot 3D streamlines from a FLASH (v.4) dataset. I've based my streamlines script off the version presented on the yt-project.org website.
I consistently get an error about going up the ghost cells in my volume. The error message is:
RuntimeError: Error: yt attempted to read outside the boundaries of a
non-periodic domain along dimension 0. Region left edge = -3e+18 code_length, Region right edge = 4.5e+18 code_length Dataset left edge = 0.0 code_length, Dataset right edge = 1.2e+19 code_length This commonly happens when trying to compute ghost cells up to the domain boundary. Two possible solutions are to load a smaller region that does not border the edge or override the periodicity for this dataset.
My FLASH datasets have x/y/zmin = 0.0 and x/y/zmax 1.2e19 cm.
I seems like yt is trying to read outside my domain, or is bumping up against the guardcells and I have no idea how to prevent it from doing this. I have tried inspecting the code for an obvious solution. A very similar error shows up when I try to make 1D profiles of the entire dataset (dd = ds.all_data()).
is it possible that yt doesn't support volumes where the (0,0,0) coordinate is a corner of the volume? Any easy solution to the above issue?
Thanks, Mikhail
yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
I don't think that (0,0,0) as a corner is a problem, but as John suggested, boundary conditions can be. A lot of yt assumes triply periodic BCs (I suspect because these are common in cosmology). You might try manually resetting the BCs in the yt data set object to periodic to see if that helps. I've done this as:
ds.periodicity = (True, True, True)
On Thu, Jul 30, 2015 at 9:34 PM, John ZuHone jzuhone@gmail.com wrote:
Hi Mikhail,
Sorry you're having trouble--any chance you can post your script using paste.yt-project.org or a similar paste service? It'll be easier to figure out what's going on then.
By the way, what are your simulation's boundary conditions?
Best,
John Z
John ZuHone Kavli Center for Astrophysics and Space Research Massachusetts Institute of Technology 77 Massachusetts Ave., 37-582G Cambridge, MA 02139 (w) 617-253-2354 (m) 781-708-5004 jzuhone@space.mit.edu jzuhone@gmail.com http://www.jzuhone.com
On Jul 30, 2015, at 9:30 PM, Mikhail Klassen klassm@mcmaster.ca wrote:
I'm trying to plot 3D streamlines from a FLASH (v.4) dataset. I've based my streamlines script off the version presented on the yt-project.org website.
I consistently get an error about going up the ghost cells in my volume. The error message is:
RuntimeError: Error: yt attempted to read outside the boundaries of a
non-periodic domain along dimension 0. Region left edge = -3e+18 code_length, Region right edge = 4.5e+18 code_length Dataset left edge = 0.0 code_length, Dataset right edge = 1.2e+19 code_length This commonly happens when trying to compute ghost cells up to the domain boundary. Two possible solutions are to load a smaller region that does not border the edge or override the periodicity for this dataset.
My FLASH datasets have x/y/zmin = 0.0 and x/y/zmax 1.2e19 cm.
I seems like yt is trying to read outside my domain, or is bumping up against the guardcells and I have no idea how to prevent it from doing this. I have tried inspecting the code for an obvious solution. A very similar error shows up when I try to make 1D profiles of the entire dataset (dd = ds.all_data()).
is it possible that yt doesn't support volumes where the (0,0,0) coordinate is a corner of the volume? Any easy solution to the above issue?
Thanks, Mikhail
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 (4)
-
John ZuHone -
Michael Zingale -
Mikhail Klassen -
Nathan Goldbaum