Welcome to sinabs-dynapcnn’s documentation!

This sinabs plugin enables porting sinabs models to chips and dev-kits with DYNAP-CNN technology.

DYNAP-CNN/Speck DevKits

• Overview of the devkits.

Table of contents

• Installation
  • Requirements
  • Installing sinabs-dynapcnn
• Getting Started With Speck/DynapCNN
  • Overview
  • Role of sinabs-dynapcnn Library
  • Chip Resources
  • Quick Start With N-MNIST
  • Visualize DVS Input
  • Play With Speck’s DVS
  • How To Leak The Neurons On The Devkit
  • Power Monitoring
  • Spike Count Visualization
  • Using Readout Layer
  • Dynapcnn Visualizer
  • Specksim
• The Basics
  • TLDR;
  • Model conversion to DYNAP-CNN core structure
  • Layer conversion
  • Parameter quantization
  • Device selection
  • List of devices currently recognized by samna
  • Placement of layers on device cores
  • Porting model to device
  • Sending and receiving spikes
  • Monitoring layer activity
• Advanced
  • Under the hood
  • The config object
  • Memory constraints
  • Attributes of interest
  • Conversion between raster and spike streams
  • Samna: The interface library to the chip
• Dangers
  • Biases/Leak are not encouraged
  • Event driven vs time-step based evaluation
  • Spike generation function
• FAQs
  • Training Tips
  • Chip Errata(Hardware Bugs)
  • Available Network Architecture
  • Available Operations
  • Device Management
  • Output Monitoring
  • Save Samna Config As Binary
• API
  • sinabs.backend.dynapcnn
• Developer’s Guide
  • How to add support for a new chip
• Contact us
  • Contact us via our public gitlab
  • Contact us via our Discord channel
• Release History
  • version 0.3

Indices and tables

• Index

• Module Index

• Search Page