rlberry.spaces.Box

class rlberry.spaces.Box(low, high, shape=None, dtype=<class 'numpy.float64'>)[source]

Bases: Box

Class that represents a space that is a cartesian product in R^n:

[a_1, b_1] x [a_2, b_2] x … x [a_n, b_n]

Inherited from gymnasium.spaces.Box for compatibility with gym.

rlberry wraps gym.spaces to make sure the seeding mechanism is unified in the library (rlberry.seeding)

Attributes:
rngnumpy.random._generator.Generator

random number generator provided by rlberry.seeding

Methods

reseed()

get new random number generator
contains(x: Any) → bool[source]

Return boolean specifying if x is a valid member of this space.

from_jsonable(sample_n: Sequence[float | int]) → list[numpy.ndarray[Any, numpy.dtype[Any]]][source]

Convert a JSONable data type to a batch of samples from this space.

is_bounded(manner: str = 'both') → bool[source]

Checks whether the box is bounded in some sense.

Args:

manner (str): One of "both", "below", "above".

Returns:

If the space is bounded

Raises:

ValueError: If manner is neither "both" nor "below" or "above"

property is_np_flattenable

Checks whether this space can be flattened to a spaces.Box.

property np_random: Generator

Lazily seed the PRNG since this is expensive and only needed if sampling from this space.

As seed() is not guaranteed to set the _np_random for particular seeds. We add a check after seed() to set a new random number generator.

reseed(seed_seq=None)[source]

Get new random number generator.

Parameters:
seed_seqnp.random.SeedSequence, rlberry.seeding.Seeder or int, defaultNone

Seed sequence from which to spawn the random number generator. If None, generate random seed. If int, use as entropy for SeedSequence. If seeder, use seeder.seed_seq

sample()[source]

Adapted from: https://raw.githubusercontent.com/openai/gym/master/gym/spaces/box.py

Generates a single random sample inside of the Box.

In creating a sample of the box, each coordinate is sampled according to the form of the interval:

  • [a, b] : uniform distribution

  • [a, oo) : shifted exponential distribution

  • (-oo, b] : shifted negative exponential distribution

  • (-oo, oo) : normal distribution

seed(seed: int | None = None) → list[int]

Seed the PRNG of this space and possibly the PRNGs of subspaces.

property shape: tuple[int, ...]

Has stricter type than gym.Space - never None.

to_jsonable(sample_n: Sequence[ndarray[Any, dtype[Any]]]) → list[list][source]

Convert a batch of samples from this space to a JSONable data type.