A deep learning library for spiking neural networks which is based on PyTorch, focuses on fast training and supports inference on neuromorphic hardware.
Currently the Merge layer merges exactly two inputs. However, there is no restriction on DynapCNN or Speck that would limit the number of inputs to a layer to 2. I suggest to make Merge more flexible.
Currently the
Merge
layer merges exactly two inputs. However, there is no restriction on DynapCNN or Speck that would limit the number of inputs to a layer to 2. I suggest to makeMerge
more flexible.