Matt, Nathan,
Thank you so much, however we would like something more adaptive, to
preserve the simulation high resolution.
Along these lines, I thought it may help to compute something like
3*v + Lap[v]*dx^2
where Lap[] is the Laplacian operator. I noticed that it is possible to
compute a gradient easily:
>>> g = d.add_gradient_fields(('deposit', 'N-BODY_density'))
>>> g
[('deposit', 'N-BODY_density_gradient_x'), ('deposit',
'N-BODY_density_gradient_y'), ('deposit', 'N-BODY_density_gradient_z'),
('deposit', 'N-BODY_density_gradient_magnitude')]
However, when I try to differentiate the gradient one more time, I get
an error:
>>> ggx = d.add_gradient_fields(('deposit', 'N-BODY_density_gradient_x'))
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/gnedin/ART/yt/yt/data_objects/static_output.py", line
1391, in add_gradient_fields
units = self.field_info[ftype, input_field].units
File "/home/gnedin/ART/yt/yt/fields/field_info_container.py", line
330, in __missing__
raise KeyError("No field named %s" % (key,))
KeyError: "No field named ('deposit', 'N-BODY_density_gradient_x')"
What am I doing wrong?
n
On 2/1/19 11:19 AM, Nathan Goldbaum wrote:
> I would also suggest looking into the arbitrary_grid data object, which
> will allow you to create a grid specifically for doing the deposit
> operation at the level of resolution you desire, decoupling the deposit
> operation from the AMR grid:
>
> Create a 20 kpc^3 grid centered on the center of the simulation, with a
> resolution of 64^3, and deposit all of the particles onto it:
>
> In [8]: grid = ds.arbitrary_grid(ds.domain_center - 10*yt.units.kpc,
> ds.domain_center + 10*yt.units.kpc, [64, 64, 64])
>
> In [9]: grid['deposit', 'all_density'].shape
> Out[9]: (64, 64, 64)
>
> On Fri, Feb 1, 2019 at 11:13 AM Matthew Turk <matthewturk(a)gmail.com
> <mailto:matthewturk@gmail.com>> wrote:
>
> Hi Nick and Hanjue,
>
> From any given selector object, you can specify the max level data
> will be drawn from; this should work with particle data in the ARTIO
> frontend, but because it may not overlap explicitly with the
> indexing system you should double check.
>
> An example:
>
> dd = ds.all_data()
> dd.max_level = 5
>
> That will restrict up to and including level 5 data.
>
> On Fri, Feb 1, 2019 at 11:11 AM Nick Gnedin <gnedin(a)fnal.gov
> <mailto:gnedin@fnal.gov>> wrote:
>
>
> Folks,
>
> We would like to use dark matter density in one of our simulations,
> however the build-in derived field ('deposit', 'N-BODY_density')
> is too
> noisy for our purposes. We can think of two ways to make it
> smoother: 1)
> to reduce the max refinement level of the underlying grid or 2) use
> SPH-like averaging on particles first before depositing them on
> the grid.
>
> Since we are newbies, could someone give us a few hints on how to
> proceed with one or both of these approaches?
>
> Many thanks,
>
> Nick
> Hanjue
> _______________________________________________
> yt-users mailing list -- yt-users(a)python.org
> <mailto:yt-users@python.org>
> To unsubscribe send an email to yt-users-leave(a)python.org
> <mailto:yt-users-leave@python.org>
>
> _______________________________________________
> yt-users mailing list -- yt-users(a)python.org
> <mailto:yt-users@python.org>
> To unsubscribe send an email to yt-users-leave(a)python.org
> <mailto:yt-users-leave@python.org>
>
