pysnn.utils¶

pysnn.utils.conv2d_output_shape(h_in, w_in, kernel_size, stride=1, padding=0, dilation=1)¶

Determine output shape after a 2D convolutional layer.

Parameters

• h_in – Height of input image.

• w_in – Width of input image.

• kernel_size – Dimensions of the convolutional kernel.

• stride – Stride of convolutional operation.

• padding – Padding of convolutional operation.

• dilation – Dilation of convolutional operation.

Returns

(Height, Width) of output image.

pysnn.utils.interspike_time(spike_array)¶

Determine number of timesteps between all consecutive spikes.

Parameters

spike_array – 1 dimensional spike array

Returns

array with the time between all consecutive spikes.

pysnn.utils.tensor_clamp(tensor, min, max)¶

Clamp tensor with a separate value for each element in the input tensor.

Parameters

• tensor – Tensor to clamp.

• min – Tensor containing minimum value for each element.

• max – Tensor containing maximum value for each element.

Returns

Clamped input Tensor.