TensorContractionLayer

Full Name

neuroptimiser.core.processes.TensorContractionLayer

Description

class TensorContractionLayer

Bases: AbstractProcess

Tensor Contraction Layer Process

This process implements a tensor contraction layer that takes a weight matrix and computes the output based on the input tensor. It is designed to be used in spiking neural network architectures where tensor contractions are required.

Inports

s_inInPort

Input port for the input tensor.

Variables

weight_matrixVar

Variable for the weight matrix used in the tensor contraction.

s_matrixVar

Variable for the output tensor after the contraction.

Outports

a_outOutPort

Output port for the output tensor after the contraction.

See also

neuroptimiser.core.models.PyTensorContractionLayerModel

Model implementation of the TensorContractionLayer process.

__init__(weights, **kwargs)

Initialise the TensorContractionLayer with the given parameters.

Parameters:

weights (np.ndarray) – Weight matrix for the tensor contraction. It should be a 2D numpy array.

Keyword Arguments:

**kwargs (dict, optional) – Additional keyword arguments to be passed to the parent class AbstractProcess.

reset()

Reset the TensorContractionLayer to its initial state.

This method is currently a placeholder and does not perform any operations for compatibility purposes.