NumpyRNG#

class deeptrack.sources.rng.NumpyRNG(n_states, seed=None)#

Bases: Source, RandomState

Class that generates multiple numpy random number generators.

It is used for creating multiple rng’s with different seeds.

Parameters#

n_states: int: The number of random number generators to create.
seed: int, optional: The seed used to initialize the first random generator. If not provided, a random seed will be generated automatically using np.random.randint().

rng: list of numpy.Random: A list of numpy.Random objects, each seeded with a unique value.

_generate_states(): list[np.random.RandomState]: Generates and returns a list of independent numpy.Random objects.
reset(): None: Resets the list of random number generators with new seeds.
__getattribute__(__name): Any: Custom attribute access to allow lazy evaluation of random number generator methods.
_create_lazy_callback(__name): callable: Creates a lazy callback for accessing methods from the numpy.Random objects.
set_index(index): self: Sets the current index and resets the random number generators.

Methods Summary

`reset`()
`set_index`(index)

Methods Documentation