|
bitrl & cuberl Documentation
Simulation engine for reinforcement learning agents
|
A scalar discrete space can be used to denote a space that only has a single value at each time. For example consider the FrozenLake environment then both the state and the action spaces are ScalarDiscreteSpace. More...
#include <space_type.h>
Public Types | |
| typedef uint_t | space_item_type |
| item_t | |
Static Public Member Functions | |
| static space_item_type | sample (bool use_end) |
| sample | |
| static space_item_type | sample (uint_t seed, bool use_end) |
| sample | |
| static std::vector< space_item_type > | sample (uint_t seed, uint_t size, bool use_end) |
| sample | |
Static Public Attributes | |
| static constexpr IntegralRange< s, e > | limits = IntegralRange<s, e>() |
| The limits of the space. | |
| static constexpr uint_t | size = e - s |
| The overall size of the space meaning how many elements can potentially the space have. | |
A scalar discrete space can be used to denote a space that only has a single value at each time. For example consider the FrozenLake environment then both the state and the action spaces are ScalarDiscreteSpace.
| typedef uint_t bitrl::envs::ScalarDiscreteSpace< s, e >::space_item_type |
item_t
|
static |
sample
|
static |
sample
| seed |
|
static |
sample
| seed |
|
staticconstexpr |
The limits of the space.
|
staticconstexpr |
The overall size of the space meaning how many elements can potentially the space have.