A deep learning library for spiking neural networks which is based on PyTorch, focuses on fast training and supports inference on neuromorphic hardware.
When trying to run the notebook cnn_to_dynapcnn.ipynb tutorial with speck2e. There is a high possibility that can't get output spikes from the devkit output buffer node.
It is because currently we need to sleep the main python process longer than we expected to get the output spikes. I discussed this with @yudi.ren, he will check the timestamp module of the FPGA is correct.
Otherwise, the current forward method of inference on the devkit might give you an empty list as output.
When trying to run the notebook
cnn_to_dynapcnn.ipynbtutorial with speck2e. There is a high possibility that can't get output spikes from the devkit output buffer node.It is because currently we need to sleep the main python process longer than we expected to get the output spikes. I discussed this with @yudi.ren, he will check the timestamp module of the FPGA is correct.
Otherwise, the current
forwardmethod of inference on the devkit might give you an empty list as output.