Neuropack is a user-friendly software-based tool that enables simulations of #neuromorphic computing with a wide range of #neuralnetwork architectures, neuron models, learning rules and #memristor settings.
Neuropack provides selectable neuron, plasticity, and device models and models memristor state changes during the training and/or the inference phases. It is a fully Python-based algo-level simulator for memristor-based neuromorphic systems.
There are two major usage scenarios: validating system concept and investigating system sensitivity.
Two related papers were presented by PhD candidate Jinqi Huang at Both papers were presented at the IEEEInternational Symposium on Circuits and Systems (ISCAS) #ISCAS22 conference earlier this year:
Paper: A tool for emulating neuromorphic architectures with memristive models and devices
Paper: Text Classification in Memristor-based Spiking Neural Networks