Spaces:
Running
Running
File size: 6,434 Bytes
375a1cf |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 |
"""Utility functions for gym spaces: batch space and iterator."""
from collections import OrderedDict
from copy import deepcopy
from functools import singledispatch
from typing import Iterator
import numpy as np
from gym.error import CustomSpaceError
from gym.spaces import Box, Dict, Discrete, MultiBinary, MultiDiscrete, Space, Tuple
BaseGymSpaces = (Box, Discrete, MultiDiscrete, MultiBinary)
_BaseGymSpaces = BaseGymSpaces
__all__ = ["BaseGymSpaces", "_BaseGymSpaces", "batch_space", "iterate"]
@singledispatch
def batch_space(space: Space, n: int = 1) -> Space:
"""Create a (batched) space, containing multiple copies of a single space.
Example::
>>> from gym.spaces import Box, Dict
>>> space = Dict({
... 'position': Box(low=0, high=1, shape=(3,), dtype=np.float32),
... 'velocity': Box(low=0, high=1, shape=(2,), dtype=np.float32)
... })
>>> batch_space(space, n=5)
Dict(position:Box(5, 3), velocity:Box(5, 2))
Args:
space: Space (e.g. the observation space) for a single environment in the vectorized environment.
n: Number of environments in the vectorized environment.
Returns:
Space (e.g. the observation space) for a batch of environments in the vectorized environment.
Raises:
ValueError: Cannot batch space that is not a valid :class:`gym.Space` instance
"""
raise ValueError(
f"Cannot batch space with type `{type(space)}`. The space must be a valid `gym.Space` instance."
)
@batch_space.register(Box)
def _batch_space_box(space, n=1):
repeats = tuple([n] + [1] * space.low.ndim)
low, high = np.tile(space.low, repeats), np.tile(space.high, repeats)
return Box(low=low, high=high, dtype=space.dtype, seed=deepcopy(space.np_random))
@batch_space.register(Discrete)
def _batch_space_discrete(space, n=1):
if space.start == 0:
return MultiDiscrete(
np.full((n,), space.n, dtype=space.dtype),
dtype=space.dtype,
seed=deepcopy(space.np_random),
)
else:
return Box(
low=space.start,
high=space.start + space.n - 1,
shape=(n,),
dtype=space.dtype,
seed=deepcopy(space.np_random),
)
@batch_space.register(MultiDiscrete)
def _batch_space_multidiscrete(space, n=1):
repeats = tuple([n] + [1] * space.nvec.ndim)
high = np.tile(space.nvec, repeats) - 1
return Box(
low=np.zeros_like(high),
high=high,
dtype=space.dtype,
seed=deepcopy(space.np_random),
)
@batch_space.register(MultiBinary)
def _batch_space_multibinary(space, n=1):
return Box(
low=0,
high=1,
shape=(n,) + space.shape,
dtype=space.dtype,
seed=deepcopy(space.np_random),
)
@batch_space.register(Tuple)
def _batch_space_tuple(space, n=1):
return Tuple(
tuple(batch_space(subspace, n=n) for subspace in space.spaces),
seed=deepcopy(space.np_random),
)
@batch_space.register(Dict)
def _batch_space_dict(space, n=1):
return Dict(
OrderedDict(
[
(key, batch_space(subspace, n=n))
for (key, subspace) in space.spaces.items()
]
),
seed=deepcopy(space.np_random),
)
@batch_space.register(Space)
def _batch_space_custom(space, n=1):
# Without deepcopy, then the space.np_random is batched_space.spaces[0].np_random
# Which is an issue if you are sampling actions of both the original space and the batched space
batched_space = Tuple(
tuple(deepcopy(space) for _ in range(n)), seed=deepcopy(space.np_random)
)
new_seeds = list(map(int, batched_space.np_random.integers(0, 1e8, n)))
batched_space.seed(new_seeds)
return batched_space
@singledispatch
def iterate(space: Space, items) -> Iterator:
"""Iterate over the elements of a (batched) space.
Example::
>>> from gym.spaces import Box, Dict
>>> space = Dict({
... 'position': Box(low=0, high=1, shape=(2, 3), dtype=np.float32),
... 'velocity': Box(low=0, high=1, shape=(2, 2), dtype=np.float32)})
>>> items = space.sample()
>>> it = iterate(space, items)
>>> next(it)
{'position': array([-0.99644893, -0.08304597, -0.7238421 ], dtype=float32),
'velocity': array([0.35848552, 0.1533453 ], dtype=float32)}
>>> next(it)
{'position': array([-0.67958736, -0.49076623, 0.38661423], dtype=float32),
'velocity': array([0.7975036 , 0.93317133], dtype=float32)}
>>> next(it)
StopIteration
Args:
space: Space to which `items` belong to.
items: Items to be iterated over.
Returns:
Iterator over the elements in `items`.
Raises:
ValueError: Space is not an instance of :class:`gym.Space`
"""
raise ValueError(
f"Space of type `{type(space)}` is not a valid `gym.Space` instance."
)
@iterate.register(Discrete)
def _iterate_discrete(space, items):
raise TypeError("Unable to iterate over a space of type `Discrete`.")
@iterate.register(Box)
@iterate.register(MultiDiscrete)
@iterate.register(MultiBinary)
def _iterate_base(space, items):
try:
return iter(items)
except TypeError:
raise TypeError(f"Unable to iterate over the following elements: {items}")
@iterate.register(Tuple)
def _iterate_tuple(space, items):
# If this is a tuple of custom subspaces only, then simply iterate over items
if all(
isinstance(subspace, Space)
and (not isinstance(subspace, BaseGymSpaces + (Tuple, Dict)))
for subspace in space.spaces
):
return iter(items)
return zip(
*[iterate(subspace, items[i]) for i, subspace in enumerate(space.spaces)]
)
@iterate.register(Dict)
def _iterate_dict(space, items):
keys, values = zip(
*[
(key, iterate(subspace, items[key]))
for key, subspace in space.spaces.items()
]
)
for item in zip(*values):
yield OrderedDict([(key, value) for (key, value) in zip(keys, item)])
@iterate.register(Space)
def _iterate_custom(space, items):
raise CustomSpaceError(
f"Unable to iterate over {items}, since {space} "
"is a custom `gym.Space` instance (i.e. not one of "
"`Box`, `Dict`, etc...)."
)
|