Legacy Tutorials

Author: fangwei123456

Because of the limited time and energy of the developers, not all tutorials can be updated along with the new version of SpikingJelly. And some tutorials have been pruned and absorbed in the new tutorials. Here we list some legacy tutorials which may be helpful.

The predecessor of Activation-based

Activation-based is called as Clock-driven in the previous version of SpikingJelly. Here is a tutorial about Clock-driven:

Clock driven

Encoders

This tutorial has not been updated. The user can refer to the old tutorial for the moment:

Clock driven: Encoder

ANN to SNN conversion

This tutorial has not been updated. The user can refer to the old tutorial for the moment:

ANN2SNN

Applications of SNNs on other tasks

Reinforcement Learning: Deep Q Learning

Reinforcement Learning: Advantage Actor Critic (A2C)

Reinforcement Learning: Proximal Policy Optimization (PPO)

Classifying Names with a Character-level Spiking LSTM

The predecessor of step mode:

Propagation Pattern

The predecessor of CUPY backend:

Accelerate with CUDA-Enhanced Neuron and Layer-by-Layer Propagation

Call for Updating

We encourage the users to update these tutorials with the master version of SpikingJelly and create the Pull Request.