explicit_tensorizer

RL4CRN.env2agent_interface.explicit_tensorizer

Explicit tensorizer.

This module defines ExplicitTensorizer, a tensorizer that converts the explicit observation produced by RL4CRN.env2agent_interface.explicit_observer.ExplicitObserver into a single flat PyTorch tensor.

The expected observation is a tuple of numpy arrays (or array-like objects), which are concatenated and converted to torch.float32 on the configured device.

ExplicitTensorizer

Bases: AbstractTensorizer

Tensorizer that flattens an explicit observation into a 1D float tensor.

__init__(device='cpu')

Initialize the tensorizer.

PARAMETER	DESCRIPTION
device	Target device for the returned tensor (e.g., 'cpu', 'cuda'). DEFAULT: 'cpu'

tensorize(observation)

Concatenate an explicit observation and convert it to a torch tensor.

The input observation is expected to be a tuple/list of array-like components (e.g., (reaction_multihot, params_cross_multihot, ...)). The components are concatenated along the last axis and returned as a single tensor.

PARAMETER	DESCRIPTION
observation	Tuple/list of numpy arrays (or array-like) representing an explicit IOCRN observation.

RETURNS	DESCRIPTION
	torch.float32 tensor on self.device containing the concatenated observation.