image_pipeline

Classes

class prt_sim.gymnasium.image_pipeline.ImagePipeline(dataset_root: Path | None = None, render_mode: str | None = None, num_image_samples: int = 1, max_steps: int = 20, device: device = device(type='cpu'))

A template Gymnasium environment that simulates an image-processing pipeline.

This environment uses a single fixed task algorithm (Yolo object detector) and the actions produce a dynamic preprocessing pipeline.

Observation:

A grayscale (H, W, 1) or RGB (H, W, 3) uint8 image

Action:

Discrete(K) - select which processing operation to apply (placeholder)

Termination:

Fixed horizon or an internal condition (customize in step)

Truncation:

Ends when a max step count is reached

Render modes:

• ‘rgb_array’ returns the current image as an np.ndarray

• ‘human’ (optional): implement if you want a GUI viewer

close() → None

There is nothing to close in this environment

get_wrapper_attr(name: str) → Any

Gets the attribute name from the environment.

has_wrapper_attr(name: str) → bool

Checks if the attribute name exists in the environment.

property np_random: Generator

Returns the environment’s internal _np_random that if not set will initialise with a random seed.

Returns:

Instances of np.random.Generator

property np_random_seed: int

Returns the environment’s internal _np_random_seed that if not set will first initialise with a random int as seed.

If np_random_seed was set directly instead of through reset() or set_np_random_through_seed(), the seed will take the value -1.

Returns:

the seed of the current np_random or -1, if the seed of the rng is unknown

Return type:

int

render() → ndarray | None

Rendering is not supported in this environment

reset(*, seed: int | None = None, options: Dict[str, Any] | None = None) → Tuple[ndarray, Dict[str, Any]]

Reset the environment to an initial state and returns an initial observation.

Parameters:

• seed (Optional[int]) – The seed that is used to initialize the environment’s random number generator

• options (Optional[Dict[str, Any]]) – Additional information to specify how the environment is reset. This is not used in this environment.

Returns:

The initial observation of the space. info (dict): A dictionary containing auxiliary information about the reset.

Return type:

observation (np.ndarray)

set_wrapper_attr(name: str, value: Any, *, force: bool = True) → bool

Sets the attribute name on the environment with value, see Wrapper.set_wrapper_attr for more info.

step(action: Dict[str, Any]) → Tuple[ndarray, float, bool, bool, Dict[str, Any]]

Run one timestep of the environment’s dynamics. When end of episode is reached, you are responsible for calling reset() to reset this environment’s state.

Parameters:

action (dict) – An action provided by the agent. This is a dictionary with keys “algorithm”: int - the index of the algorithm to apply “parameters”: np.ndarray - the parameters for the algorithm scaled between 0 and 1

Returns:

Agent’s observation of the current environment reward (float): Amount of reward returned after previous action terminated (bool): Whether the episode has ended. Further step() calls will return undefined results truncated (bool): Whether the episode was truncated (max steps reached). Further step() calls will return undefined results info (dict): Contains auxiliary diagnostic information (helpful for debugging, and sometimes learning)

Return type:

observation (np.ndarray)

property unwrapped: Env[ObsType, ActType]

Returns the base non-wrapped environment.

Returns:

The base non-wrapped gymnasium.Env instance

Return type:

Env