Resize

class torchrl.envs.transforms.Resize(w: int, h: int, interpolation: str = 'bilinear', in_keys: Optional[Sequence[Union[str, Tuple[str, ...]]]] = None, out_keys: Optional[Sequence[Union[str, Tuple[str, ...]]]] = None)

Resizes a pixel observation.

Parameters:

• w (int) – resulting width

• h (int) – resulting height

• interpolation (str) – interpolation method

transform_observation_spec(observation_spec: TensorSpec) → TensorSpec

Transforms the observation spec such that the resulting spec matches transform mapping.

Parameters:

observation_spec (TensorSpec) – spec before the transform

Returns:

expected spec after the transform