A deep learning library for spiking neural networks which is based on PyTorch, focuses on fast training and supports inference on neuromorphic hardware.
[x] Tests for the changes have been added (for bug fixes/features)
[x] Docs have been added / updated (for bug fixes / features)
[x] I have performed a self-review of my code
[x] Will this be part of a product update? If yes, please write one line about this on the CHANGELOG.md
What kind of change does this PR introduce? (Bug fix, feature, docs update, ...)
Minor feature upgrade: More flexibility in DynapcnnVisualizer class.
What is the current behavior? (You can also link to an open issue here)
Spike count plots are always included in the DynapcnnVisualizer, which can be annoying (see #238 )
Readout predictions are based on a logic pre-defined in samna, which is sometimes not adequate to the task.
What is the new behavior (if this is a feature change)?
Spike count plots of the DynapcnnVisualizer are optional now.
Custom JIT filters can be passed as readout nodes to the DynapcnnVisualizer class. This makes it possible to use custom logic for making predictions in the readout plot.
Does this PR introduce a breaking change? (What changes might users need to make in their application due to this PR?)
No, just added arguments. Existing code will keep working.
Checklist before requesting a review
[x] Tests for the changes have been added (for bug fixes/features)
[x] Docs have been added / updated (for bug fixes / features)
[x] I have performed a self-review of my code
[x] Will this be part of a product update? If yes, please write one line about this on the CHANGELOG.md
What kind of change does this PR introduce? (Bug fix, feature, docs update, ...) Minor feature upgrade: More flexibility in
DynapcnnVisualizerclass.What is the current behavior? (You can also link to an open issue here)
Spike count plots are always included in the
DynapcnnVisualizer, which can be annoying (see #238 )Readout predictions are based on a logic pre-defined in samna, which is sometimes not adequate to the task.
What is the new behavior (if this is a feature change)?
Spike count plots of the
DynapcnnVisualizerare optional now.Custom JIT filters can be passed as readout nodes to the
DynapcnnVisualizerclass. This makes it possible to use custom logic for making predictions in the readout plot.Does this PR introduce a breaking change? (What changes might users need to make in their application due to this PR?) No, just added arguments. Existing code will keep working.
Other information: