A deep learning library for spiking neural networks which is based on PyTorch, focuses on fast training and supports inference on neuromorphic hardware.
The DynapcnnVisualizer accepts and argument at instantiation named readout_prediction_threshold. This value is used nowhere. The argument should either be removed, or a different "majority readout node" has to be used, which makes use of the prediction threshold. Furthermore the documentation (in code and here is wrong and has to be updated accordingly.
bauerfe
commented
2 weeks ago
The DynapcnnVisualizer accepts and argument at instantiation named
readout_prediction_threshold. This value is used nowhere. The argument should either be removed, or a different "majority readout node" has to be used, which makes use of the prediction threshold. Furthermore the documentation (in code and here is wrong and has to be updated accordingly.