Getting Started With Speck/DynapCNN# Overview DYNAP-CNN and SPECK Role of sinabs-dynapcnn Library Chip Resources Key Features Quick Start With N-MNIST Data Preparation CNN-To-SNN Train SNN with BPTT Deploy SNN To The Devkit In-depth Investigation Of SNN Deployment Stage Visualize DVS Input 1. Open devkit 2. Build samna graph 3. Lauch visualizer window 4. Launch the devkit and start visualization 5. Stop the devkit and visualization Play With Speck’s DVS Monitor DVS Events Merge Polarity Not Merge Polarity Pooling Cropping Mirroring Disable DVS Events From Entering The Processor Output Destination Layer Select Switching On/Off Polarity. Hot-Pixel Filtering How To Leak The Neurons On The Devkit Power Monitoring 1. Power consumption of the devkit 2. Monitor the idle power 3. Monitor the dynamic power Spike Count Visualization 1. Create a 1-layer CNN which can be deployed to the devkit 2. Prepare for deployment 3. Start Visualization Using Readout Layer What is readout layer 1. Create a 1-layer CNN which can be deployed to the devkit 2. Two different types of output events from readout layer 3. Prepare for deployment 4. Start readout from devkit Dynapcnn Visualizer Introduction Available plots in samnagui Useful samna nodes for visualizing and Just-In-Time (JIT) compiled nodes. Dynapcnn Visualizer Specksim Introduction Setup Supported architecture Supported layers Input/Output How-to-use Drawbacks and possible questions Try it yourself
