FAQs

• Troubleshooting and Tips
  • Memory Constrains On The Hardware
  • SynOps/s Bandwidth Constrains On The Hardware
  • The Reset Mechanism Of Neuron Membrane Potential
  • Weight Clipping During Training
  • Selection Of Spike Function
• Chip Errata(Hardware Bugs)
  • Speck2e/2f
• Available Network Architecture
  • Brief Introduction of the DynapCNN Core
  • What network structure can I define?
  • How to make use of the graph extraction feature?
  • Can I achieve a “Residual Connection” like ResNet does?
  • How can I define “Residual Connection” manually?
  • What execution order should I be aware of when I am implementing a sequential structure?
  • Memory Constrains
  • Feature Map Size Constrains
  • Limitation of Using ReadoutLayer
• Available Operations
  • Available torch.nn.modules for Devkits
  • Available sinabs.layers for Devkits
  • Neuron Leakage
• Device Management
  • How Do I List All Connected Devices And Their IDs?
  • How Do I Open The Device
  • How Do I Reset The Device
  • How Do I Close The Device
• Output Monitoring
  • How to Read The Output from Hidden Layers.
  • How to Read The Vmem of the Neurons
• Save Samna Config As Binary
• How to add support for a new chip
• How To Leak The Neurons On The Devkit