Loading Generic Particle Data¶
This example creates a fake in-memory particle dataset and then loads it as a yt dataset using the load_particles function.
Our “fake” dataset will be numpy arrays filled with normally distributed randoml particle positions and uniform particle masses. Since real data is often scaled, I arbitrarily multiply by 1e6 to show how to deal with scaled data.
[1]:
import numpy as np
rng = np.random.default_rng()
n_particles = 5_000_000
ppx, ppy, ppz = 1e6 * rng.normal(size=(3, n_particles))
ppm = np.ones(n_particles)
The load_particles function accepts a dictionary populated with particle data fields loaded in memory as numpy arrays or python lists:
[2]:
data = {
("io", "particle_position_x"): ppx,
("io", "particle_position_y"): ppy,
("io", "particle_position_z"): ppz,
("io", "particle_mass"): ppm,
}
To hook up with yt’s internal field system, the dictionary keys must be ‘particle_position_x’, ‘particle_position_y’, ‘particle_position_z’, and ‘particle_mass’, as well as any other particle field provided by one of the particle frontends.
The load_particles function transforms the data dictionary into an in-memory yt Dataset object, providing an interface for further analysis with yt. The example below illustrates how to load the data dictionary we created above.
[3]:
import yt
from yt.units import Msun, parsec
bbox = 1.1 * np.array(
[[min(ppx), max(ppx)], [min(ppy), max(ppy)], [min(ppz), max(ppz)]]
)
ds = yt.load_particles(data, length_unit=1.0 * parsec, mass_unit=1e8 * Msun, bbox=bbox)
The length_unit and mass_unit are the conversion from the units used in the data dictionary to CGS. I’ve arbitrarily chosen one parsec and 10^8 Msun for this example.
The n_ref parameter controls how many particle it takes to accumulate in an oct-tree cell to trigger refinement. Larger n_ref will decrease poisson noise at the cost of resolution in the octree.
Finally, the bbox parameter is a bounding box in the units of the dataset that contains all of the particles. This is used to set the size of the base octree block.
This new dataset acts like any other yt Dataset object, and can be used to create data objects and query for yt fields.
[4]:
ad = ds.all_data()
print(ad.mean(("io", "particle_position_x")))
print(ad.sum(("io", "particle_mass")))
238.7069071770326 code_length
9.94207930000004e+47 g
We can project the particle mass field like so:
[5]:
prj = yt.ParticleProjectionPlot(ds, "z", ("io", "particle_mass"))
prj.set_width((8, "Mpc"))
[5]:
Finally, one can specify multiple particle types in the data directory by setting the field names to be field tuples (the default field type for particles is "io") if one is not specified:
[6]:
n_gas_particles = 1_000_000
n_star_particles = 1_000_000
n_dm_particles = 2_000_000
ppxg, ppyg, ppzg = 1e6 * rng.normal(size=(3, n_gas_particles))
ppmg = np.ones(n_gas_particles)
hsml = 10000 * np.ones(n_gas_particles)
dens = 2.0e-4 * np.ones(n_gas_particles)
ppxd, ppyd, ppzd = 1e6 * rng.normal(size=(3, n_dm_particles))
ppmd = np.ones(n_dm_particles)
ppxs, ppys, ppzs = 5e5 * rng.normal(size=(3, n_star_particles))
ppms = 0.1 * np.ones(n_star_particles)
bbox = 1.1 * np.array(
[
[
min(ppxg.min(), ppxd.min(), ppxs.min()),
max(ppxg.max(), ppxd.max(), ppxs.max()),
],
[
min(ppyg.min(), ppyd.min(), ppys.min()),
max(ppyg.max(), ppyd.max(), ppys.max()),
],
[
min(ppzg.min(), ppzd.min(), ppzs.min()),
max(ppzg.max(), ppzd.max(), ppzs.max()),
],
]
)
data2 = {
("gas", "particle_position_x"): ppxg,
("gas", "particle_position_y"): ppyg,
("gas", "particle_position_z"): ppzg,
("gas", "particle_mass"): ppmg,
("gas", "smoothing_length"): hsml,
("gas", "density"): dens,
("dm", "particle_position_x"): ppxd,
("dm", "particle_position_y"): ppyd,
("dm", "particle_position_z"): ppzd,
("dm", "particle_mass"): ppmd,
("star", "particle_position_x"): ppxs,
("star", "particle_position_y"): ppys,
("star", "particle_position_z"): ppzs,
("star", "particle_mass"): ppms,
}
ds2 = yt.load_particles(
data2, length_unit=1.0 * parsec, mass_unit=1e8 * Msun, bbox=bbox
)
We now have separate "gas", "dm", and "star" particles. Since the "gas" particles have "density" and "smoothing_length" fields, they are recognized as SPH particles:
[7]:
ad = ds2.all_data()
c = np.array([ad.mean(("gas", ax)).to("code_length") for ax in "xyz"])
[8]:
slc = yt.SlicePlot(ds2, "z", ("gas", "density"), center=c)
slc.set_zlim(("gas", "density"), 1e-19, 2.0e-18)
slc.set_width((4, "Mpc"))
slc.show()
---------------------------------------------------------------------------
UnitConversionError Traceback (most recent call last)
Cell In[8], line 1
----> 1 slc = yt.SlicePlot(ds2, "z", ("gas", "density"), center=c)
2 slc.set_zlim(("gas", "density"), 1e-19, 2.0e-18)
3 slc.set_width((4, "Mpc"))
File /var/jenkins_home/workspace/yt_docs/yt/visualization/plot_window.py:1842, in AxisAlignedSlicePlot.__init__(self, ds, normal, fields, center, width, axes_unit, origin, fontsize, field_parameters, window_size, aspect, data_source, buff_size, north_vector)
1840 slc.get_data(fields)
1841 validate_mesh_fields(slc, fields)
-> 1842 PWViewerMPL.__init__(
1843 self,
1844 slc,
1845 bounds,
1846 origin=origin,
1847 fontsize=fontsize,
1848 fields=fields,
1849 window_size=window_size,
1850 aspect=aspect,
1851 buff_size=buff_size,
1852 geometry=ds.geometry,
1853 )
1854 if axes_unit is None:
1855 axes_unit = get_axes_unit(width, ds)
File /var/jenkins_home/workspace/yt_docs/yt/visualization/plot_window.py:864, in PWViewerMPL.__init__(self, *args, **kwargs)
862 self._plot_type = kwargs.pop("plot_type")
863 self._splat_color = kwargs.pop("splat_color", None)
--> 864 PlotWindow.__init__(self, *args, **kwargs)
866 # import type here to avoid import cycles
867 # note that this import statement is actually crucial at runtime:
868 # the filter methods for the present class are defined only when
869 # fixed_resolution_filters is imported, so we need to guarantee
870 # that it happens no later than instantiation
871 from yt.visualization.plot_modifications import PlotCallback
File /var/jenkins_home/workspace/yt_docs/yt/visualization/plot_window.py:252, in PlotWindow.__init__(self, data_source, bounds, buff_size, antialias, periodic, origin, oblique, window_size, fields, fontsize, aspect, setup, geometry)
249 self._projection = get_mpl_transform(projection)
250 self._transform = get_mpl_transform(transform)
--> 252 self._setup_plots()
254 for field in self.data_source._determine_fields(self.fields):
255 finfo = self.data_source.ds._get_field_info(field)
File /var/jenkins_home/workspace/yt_docs/yt/visualization/plot_window.py:1066, in PWViewerMPL._setup_plots(self)
1062 extenty.convert_to_units(unit_y)
1064 extent = [*extentx, *extenty]
-> 1066 image = self.frb.get_image(f)
1067 mask = self.frb.get_mask(f)
1068 assert mask is None or mask.dtype == bool
File /var/jenkins_home/workspace/yt_docs/yt/visualization/fixed_resolution.py:213, in FixedResolutionBuffer.get_image(self, key)
211 def get_image(self, key, /) -> ImageArray:
212 if not (key in self.data and self._data_valid):
--> 213 self._generate_image_and_mask(key)
214 return self.data[key]
File /var/jenkins_home/workspace/yt_docs/yt/visualization/fixed_resolution.py:179, in FixedResolutionBuffer._generate_image_and_mask(self, item)
176 b = float(b.in_units("code_length"))
177 bounds.append(b)
--> 179 buff, mask = self.ds.coordinates.pixelize(
180 self.data_source.axis,
181 self.data_source,
182 item,
183 bounds,
184 self.buff_size,
185 int(self.antialias),
186 return_mask=True,
187 )
189 buff = self._apply_filters(buff)
191 # FIXME FIXME FIXME we shouldn't need to do this for projections
192 # but that will require fixing data object access for particle
193 # projections
File /var/jenkins_home/workspace/yt_docs/yt/geometry/coordinates/cartesian_coordinates.py:237, in CartesianCoordinateHandler.pixelize(self, dimension, data_source, field, bounds, size, antialias, periodic, return_mask)
234 mask = np.squeeze(np.transpose(mask, (yax, xax, ax)))
236 elif self.axis_id.get(dimension, dimension) is not None:
--> 237 buff, mask = self._ortho_pixelize(
238 data_source, field, bounds, size, antialias, dimension, periodic
239 )
240 else:
241 buff, mask = self._oblique_pixelize(
242 data_source, field, bounds, size, antialias
243 )
File /var/jenkins_home/workspace/yt_docs/yt/geometry/coordinates/cartesian_coordinates.py:504, in CartesianCoordinateHandler._ortho_pixelize(self, data_source, field, bounds, size, antialias, dim, periodic)
494 buff_den = np.zeros(size, dtype="float64")
496 for chunk in data_source.chunks([], "io"):
497 pixelize_sph_kernel_slice(
498 buff,
499 mask_uint8,
500 chunk[ptype, px_name].to("code_length"),
501 chunk[ptype, py_name].to("code_length"),
502 chunk[ptype, "smoothing_length"].to("code_length"),
503 chunk[ptype, "mass"].to("code_mass"),
--> 504 chunk[ptype, "density"].to("code_density"),
505 chunk[field].in_units(ounits),
506 bnds,
507 check_period=int(periodic),
508 period=period,
509 )
510 if normalize:
511 pixelize_sph_kernel_slice(
512 buff_den,
513 mask_uint8,
(...)
522 period=period,
523 )
File /var/jenkins_home/workspace/yt_docs/.venv/lib/python3.11/site-packages/unyt/array.py:971, in unyt_array.to(self, units, equivalence, **kwargs)
929 def to(self, units, equivalence=None, **kwargs):
930 """
931 Creates a copy of this array with the data converted to the
932 supplied units, and returns it.
(...)
969 898755178736817.6 J
970 """
--> 971 return self.in_units(units, equivalence=equivalence, **kwargs)
File /var/jenkins_home/workspace/yt_docs/.venv/lib/python3.11/site-packages/unyt/array.py:899, in unyt_array.in_units(self, units, equivalence, **kwargs)
897 else:
898 new_units = units
--> 899 (conversion_factor, offset) = self.units.get_conversion_factor(
900 new_units, self.dtype
901 )
902 dsize = max(2, self.dtype.itemsize)
903 if self.dtype.kind in ("u", "i"):
File /var/jenkins_home/workspace/yt_docs/.venv/lib/python3.11/site-packages/unyt/unit_object.py:676, in Unit.get_conversion_factor(self, other_units, dtype)
649 def get_conversion_factor(self, other_units, dtype=None):
650 """Get the conversion factor and offset (if any) from one unit
651 to another
652
(...)
674 (1.7999999999999998, -31.999999999999886)
675 """
--> 676 return _get_conversion_factor(self, other_units, dtype)
File /var/jenkins_home/workspace/yt_docs/.venv/lib/python3.11/site-packages/unyt/unit_object.py:921, in _get_conversion_factor(old_units, new_units, dtype)
898 """
899 Get the conversion factor between two units of equivalent dimensions. This
900 is the number you multiply data by to convert from values in `old_units` to
(...)
918
919 """
920 if old_units.dimensions != new_units.dimensions:
--> 921 raise UnitConversionError(
922 old_units, old_units.dimensions, new_units, new_units.dimensions
923 )
924 old_basevalue = old_units.base_value
925 old_baseoffset = old_units.base_offset
UnitConversionError: Cannot convert between 'dimensionless' (dim '1') and 'code_density' (dim '(mass)/(length)**3').