[yt-users] Re: Error making a movie with RAMSES data

I see. I will try this. However, I noticed another issue with the 'data' where printing it gives me a zero array

def star_filter(pfilter,data):

age = data["all", "particle_age"]

filter=age.in_units('Myr')>0

return filter

data={'star_density':dd[('deposit', 'new_stars_density')]} print(data)

I tried replacing the 'deposit' field with a particle field ('particle_position_x') and I get non-zero values. Any suggestions in this issue? I am basically trying to make a movie of new star particles spawning and disappearing after certain time and appreciate if there's any simpler alternative in yt to do this. Thanks! On Tue, Jan 8, 2019 at 6:52 PM Nathan Goldbaum wrote:

Did you cut off the error message? I think there are more lines.

The problem here is that "data" should be a dictionary mapping a field name to a 3D array:

data = {'star_density': dd[('deposit', 'new_stars_density')]}

That should get past the error you are currently seeing.

You will then hit another error because the MeshSource is only appropriate to use with unstructured mesh data.

Take a look at the volume rendering documentation for more detail about how to use the volume renderer:

http://yt-project.org/doc/visualizing/volume_rendering.html

In particular you should be able to use yt.create_scene with the reloaded dataset.

-Nathan

On Tue, Jan 8, 2019 at 11:45 AM Vadlamani Samhitha < vadlamani.samhitha@gmail.com> wrote:

...

Thanks for the suggestion. However, this is how I'm doing it, but not sure if this is the right way to proceed as I end up with an error:

...

filename= "/lunarc/nobackup/users/samvad/FINAL-50-0.5/output/output_00051/info_00051.txt"

ds=yt.load(filename

ad=ds.all_data()

...

dd=ds.arbitrary_grid(left_edge=[0,0,0], right_edge=[1,1,1], dims=[512,512,512])

data=dd[('deposit', 'new_stars_density')]

bbox=np.array([[0,1],[0,1],[0,1]])

ds=yt.load_uniform_grid(data, data.shape, bbox)

ms =MeshSource(ds, ('deposit', 'new_stars_density'))

ERROR:

Traceback (most recent call last):

File "movie_new_stars.py", line 36, in <module>

ds=yt.load_uniform_grid(data, data.shape, bbox)

File "/home/samvad/yt-conda/lib/python3.6/site-packages/yt/frontends/stream/data_structures.py", line 666, in load_uniform_grid

data, grid_dims=tuple(domain_dimensions))

File "/home/samvad/yt-conda/lib/python3.6/site-packages/yt/frontends/stream/data_structures.py", line 477, in process_data for field, val in data.items(): Thanks!

On Mon, Jan 7, 2019 at 7:27 PM Nathan Goldbaum wrote:

...

This failure is happening because the MeshSource only works with unstructured mesh data. It will not work with your RAMSES AMR data.

Your best bet for volume rendering RAMSES data is to first deposit the data onto a uniform resolution mesh, reaload that data using yt.load_uniform_grid, and volume render the reloaded dataset.

We currently do not support volume rendering RAMSES data. We realize this is something that a lot of users want but unfortunately do not have funded developer time right now for working on fixing that situation.

Apologies for the trouble, hopefully this makes more sense now.

-Nathan

On Mon, Jan 7, 2019 at 4:30 AM Vadlamani Samhitha < vadlamani.samhitha@gmail.com> wrote:

...

Hi, I'm trying to make a movie by depositing particles on to the mesh:

...

import yt

from yt.visualization.volume_rendering.api import MeshSource

from yt.units import Myr

import yt.visualization.eps_writer as eps

from yt.data_objects.particle_filters import add_particle_filter

import pylab as plt

plt.switch_backend('agg')

def star_filter(pfilter,data):

age = data["all", "particle_age"]

filter = age.in_units('Myr') > 0

return filter

add_particle_filter("new_stars", function=star_filter, filtered_type= "all", requires=["particle_age"])

ini=19

fin=21

CMAP = 'hot'

VMIN = 1e3

VMAX = 1e10

for x in range(ini, fin):

filename= "/lunarc/nobackup/users/samvad/FINAL-50-0.5/output/output_%0*d/info_%0*d.txt" % (5,x,5,x)

ds=yt.load(filename)

time = ds.current_time

ds.add_particle_filter("new_stars")

ms =MeshSource(ds, ('deposit', 'new_stars_density'))

sc.add_camera(ds)

camera_position = ds.arr([0, 0, 0], 'code_length')

cam.focus = ds.domain_center

north_vector = ds.arr([0, 0, 0], 'dimensionless')

cam.width = ds.arr([ 400, 400, 400], 'kpc')

cam.resolution = (800, 800)

...

cam.set_position(camera_position, north_vector)

# actually make the image here

im = ms.render(cam, cmap=CMAP, color_bounds=(VMIN, VMAX))

plt.clf()

fig = plt.gcf()

ax = plt.gca()

ax.imshow(im, interpolation='nearest', origin='lower')

p = ax.imshow(ms.data, visible=False, cmap=CMAP, vmin=VMIN, vmax=VMAX)

cb = fig.colorbar(p)

cb.set_label(field_name[1])

ax.text(25, 750, 'time = %.2e' % time, color='k')

...

ax.axes.get_xaxis().set_visible(False)

plt.savefig('movie_frames/test_%.3d' % x)

But ending up with error:

Traceback (most recent call last):

File "movie_new_stars.py", line 32, in <module>

ms =MeshSource(ds, ('deposit', 'new_stars_density'))

File "/home/samvad/yt-conda/lib/python3.6/site-packages/yt/visualization/volume_rendering/render_source.py", line 536, in __init__

self.build_volume_embree()

File "/home/samvad/yt-conda/lib/python3.6/site-packages/yt/visualization/volume_rendering/render_source.py", line 593, in build_volume_embree

mesh_id = int(ftype[-1]) - 1

...

ValueError: invalid literal for int() with base 10: 't'

Any leads on this are highly appreciated.

...

_______________________________________________ yt-users mailing list -- yt-users@python.org To unsubscribe send an email to yt-users-leave@python.org

_______________________________________________ yt-users mailing list -- yt-users@python.org To unsubscribe send an email to yt-users-leave@python.org

_______________________________________________ yt-users mailing list -- yt-users@python.org To unsubscribe send an email to yt-users-leave@python.org

_______________________________________________ yt-users mailing list -- yt-users@python.org To unsubscribe send an email to yt-users-leave@python.org