Skip to content



If you want to run the code below, consider downloading the demo data or use the TNGLab online.

Loading data with units

Loading data sets with

from scida import load
ds = load("./snapdir_030")

will automatically attach units to the data. This can be deactivated by passing "units=False" to the load function. By default, code units are used, alternatively, cgs conversions can be applied by passing "units='cgs'" (experimental).

Units are introduced via the pint package, see there for more details.

Sometimes, the units cannot be inferred or parsed. This will be indicated in the output following the call to load(), e.g. as:

Missing units for 1 fields.
Fields with missing units:
  - /PartType0/FieldWithoutUnits (missing)

You can obtain more information on the cause by setting

>>> import logging
>>> logging.getLogger().setLevel(logging.DEBUG)

before calling load().

Units for custom datasets can also be manually be specified using unit files, see here.

Using data with units

>>> gas =["PartType0"]
>>> gas["Coordinates"]
dask.array<mul, shape=(18540104, 3), dtype=float64, chunksize=(5592405, 3), chunktype=numpy.ndarray> <Unit('code_length')>

We can access the underlying dask array and the units separately:

>>> gas["Coordinates"].magnitude, gas["Coordinates"].units
(dask.array<mul, shape=(18540104, 3), dtype=float64, chunksize=(5592405, 3), chunktype=numpy.ndarray>,

Unit conversions

We can change units for evaluation as desired:

>>> coords = gas["Coordinates"]
>>> # here the default system is cgs, thus we get the same result from
>>> coords.to_base_units()
dask.array<mul, shape=(18540104, 3), dtype=float64, chunksize=(5592405, 3), chunktype=numpy.ndarray> <Unit('centimeter')>

The unit registry

The unit registry keeps all units. There is no global registry, but each dataset has its own registry as attribute ureg. The use of a global registry (or lack thereof here) can lead to some confusion, please consult the pint documentation when in doubt.

>>> ureg = ds.ureg
>>> # get the unit meter
>>> ureg("m")
1 <Unit('meter')>
>>> # define an array with units meter (dask arrays analogously)
>>> import numpy as np
>>> np.arange(10) * ureg("m")
array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9]) <Unit('meter')>

Synthesize new dask arrays with units

>>> energy_restframe = (gas["Masses"]*ureg("c")**2).to("erg")  # E=mc^2
>>> energy_restframe
dask.array<mul, shape=(18540104,), dtype=float64, chunksize=(18540104,), chunktype=numpy.ndarray> <Unit('erg')>

Custom units

>>> ureg.define("halfmeter = 0.5 * m")
>>> # first particle coordinates in halfmeters
array([6.64064027e+23, 2.23858253e+24, 1.94176712e+24]) <Unit('halfmeter')>