discretize

discretize#

This module defines methods for discretizing convoluitional torch layers and SINABS spiking layers.

sinabs.backend.dynapcnn.discretize.determine_discretization_scale(obj: Tensor, bit_precision: int) → float[source]#

Determine a scale for discretization.

Determine how much the values of a torch tensor can be scaled in order to fit the given precision

Parameters:

  • obj (torch.Tensor) – Tensor that is to be scaled.

  • bit_precision (int) – The precision in bits.

Returns:

The scaling factor

Return type:

float

sinabs.backend.dynapcnn.discretize.discretize_conv(layer: Conv2d, spk_thr: float, spk_thr_low: float, spk_state: Tensor | None = None, to_int: bool = True) → Conv2d[source]#

Discretize convolutional layer.

This function takes a 2D convolutional layer and parameters of a subsequent spiking layer to return a discretized copy of the convolutional layer.

Parameters:

  • layer (nn.Conv2d) – Convolutional layer.

  • spk_thr (float) – Upper threshold of subsequent spiking layer.

  • spk_thr_low (float) – Lower threshold of subsequent spiking layer.

  • spk_state (torch.Tensor) – State of spiking layer. Defaults to None.

  • to_int (bool) – Use integer types for discretized parameter. Defaults to True.

Returns:

Discretized copy of convolutional layer

Return type:

Conv2d

sinabs.backend.dynapcnn.discretize.discretize_conv_(layer: Conv2d, spk_thr: float, spk_thr_low: float, spk_state: Tensor | None = None, to_int: bool = True) → Conv2d[source]#

Discretize convolutional layer, in-place.

This function discretizes a 2D convolutional layer in-place, based on parameters of a subsequent spiking layer.

Parameters:

  • layer (nn.Conv2d) – Convolutional layer.

  • spk_thr (float) – Upper threshold of subsequent spiking layer.

  • spk_thr_low (float) – Lower threshold of subsequent spiking layer.

  • spk_state (torch.Tensor) – State of spiking layer. Defaults to None.

  • to_int (bool) – Use integer types for discretized parameter. Defaults to True.

Returns:

Discretized convolutional layer

Return type:

Conv2d

sinabs.backend.dynapcnn.discretize.discretize_conv_spike(conv_lyr: Conv2d, spike_lyr: IAF, to_int: bool = True) → Tuple[Conv2d, IAF][source]#

Discretize convolutional and spiking layers together.

This function takes a 2D convolutional and a spiking layer and returns a copy of each, with discretized weights, bias and threshold.

Parameters:

  • conv_lyr (nn.Conv2d) – Convolutional layer.

  • spike_lyr (sl.IAF) – Spiking layer.

  • to_int (bool) – Use integer types for discretized parameter.

Returns:

Tuple containing a discretized copy of convolutional layer and a discretized copy of spiking layer.

Return type:

Tuple[Conv2d, IAF]

sinabs.backend.dynapcnn.discretize.discretize_conv_spike_(conv_lyr: Conv2d, spike_lyr: IAF, to_int: bool = True) → Tuple[Conv2d, IAF][source]#

Discretize convolutional and spiking layers together, in-place.

This function takes a 2D convolutional and a spiking layer and discretizes weights, bias and threshold in-place.

Parameters:

  • conv_lyr (nn.Conv2d) – Convolutional layer.

  • spike_lyr (sl.IAF) – Spiking layer.

  • to_int (bool) – Use integer types for discretized parameter.

Returns:

A tuple containing a discretized convolutional layer and a discretized spiking layer.

Return type:

Tuple[Conv2d, IAF]

sinabs.backend.dynapcnn.discretize.discretize_scalar(obj: float, scaling: float) → int[source]#

Scale a float and cast it to discrete integer values.

Parameters:

  • obj (float) – Value that is to be discretized.

  • scaling (float) – Scaling factor to be applied before discretization.

Returns:

Scaled and discretized copy of obj.

Return type:

int

sinabs.backend.dynapcnn.discretize.discretize_spk(layer: IAF, conv_weight: Tensor, conv_bias: Tensor | None = None, to_int: bool = True) → IAF[source]#

Discretize spiking layer.

This function takes a spiking layer and parameters of a preceding convolutional layer to return a discretized copy of the spiking layer.

Args: layer (sl.IAF): Spiking layer. conv_weight (torch.Tensor): Weight tensor of preceding convolutional layer. conv_bias (torch.Tensor): Bias of preceding convolutional layer.

Optional argument, defaults to None.

to_int (bool): Use integer types for discretized parameter.

Returns:

Discretized copy of spiking layer

Parameters:

  • layer (IAF)

  • conv_weight (Tensor)

  • conv_bias (Tensor | None)

  • to_int (bool)

Return type:

IAF

sinabs.backend.dynapcnn.discretize.discretize_spk_(layer: IAF, conv_weight: Tensor, conv_bias: Tensor | None = None, to_int: bool = True) → IAF[source]#

Discretize spiking layer in-place.

This function discretizes a spiking layer in-place, based on parameters of a preceding convolutional layer.

Parameters:

  • layer (sl.IAF) – Spiking layer.

  • conv_weight (torch.Tensor) – Weight tensor of preceding convolutional layer.

  • conv_bias (torch.Tensor) – Bias of preceding convolutional layer. Optional argument, defaults to None.

  • to_int (bool) – Use integer types for discretized parameter.

Returns:

Discretized spiking

Return type:

IAF

sinabs.backend.dynapcnn.discretize.discretize_tensor(obj: Tensor, scaling: float, to_int: bool = True) → Tensor[source]#

Scale a torch.Tensor and cast it to discrete integer values.

Parameters:

  • obj (torch.Tensor) – Tensor that is to be discretized.

  • scaling (float) – Scaling factor to be applied before discretization.

  • to_int (bool) – If False, round the values, but don’t cast to Int. Defaults to True.

Returns:

Scaled and discretized copy of obj.

Return type:

Tensor

Contents