A deep learning library for spiking neural networks.

Test status chat on Discord DOI

Norse aims to exploit the advantages of bio-inspired neural components, which are sparse and event-driven - a fundamental difference from artificial neural networks. Norse expands PyTorch with primitives for bio-inspired neural components, bringing you two advantages: a modern and proven infrastructure based on PyTorch and deep learning-compatible spiking neural network components.

Documentation: norse.github.io/norse/

Getting started

To try Norse, the best option is to run one of the jupyter notebooks on Google collab.

Alternatively, you can install Norse and run one of the included tasks such as MNIST:

python -m norse.task.mnist

The Quickstart and Working with Norse pages show how to build your own models with Norse while explaining a few fundamental concepts around spiking neural networks.

Installing Norse

Note that we assume you are using Python version 3.7+, are in a terminal friendly environment, and have installed the necessary requirements, depending on your installation method. More detailed installation instructions are available here: Installing Norse.

From PyPi
pip install norse
From source
pip install -qU git+https://github.com/norse/norse
Pip, PyTorch
With Docker
docker pull quay.io/norse/norse
From Conda
conda install -c norse norse
Anaconda or Miniconda

Running examples

Norse is bundled with a number of example tasks, serving as short, self contained, correct examples (SSCCE). They can be run by invoking the norse module from the base directory. More information and tasks are available in our documentation and in your console by typing: python -m norse.task.<task> --help, where <task> is one of the task names.

  • To train an MNIST classification network, invoke

    python -m norse.task.mnist
  • To train a CIFAR classification network, invoke

    python -m norse.task.cifar10
  • To train the cartpole balancing task with Policy gradient, invoke

    python -m norse.task.cartpole

Norse is compatible with PyTorch Lightning, as demonstrated in the PyTorch Lightning MNIST task variant (requires PyTorch lightning):

python -m norse.task.mnist_pl --gpus=4

Read more in our Introduction to spiking systems and visit our Jupyter Notebook examples.

Advanced uses and opimizations

Norse is meant to be used as a library for spiking neural networks in customized deep learning models. This typically means porting other models to the spiking/temporal domain, extending existing models, or starting completely from scratch. All three use cases are motivated and briefly described in Working with Norse.

Unfortunately, spiking neural networks are resource intensive. The page on Hardware acceleration explains how to accelerate the simulations using dedicated hardware.


Contributions are warmly encouraged and always welcome. However, we also have high expectations around the code base so if you wish to contribute, please refer to our contribution guidelines.


Norse is created by

More information about Norse can be found in our documentation. The research has received funding from the EC Horizon 2020 Framework Programme under Grant Agreements 785907 and 945539 (HBP) and by the Deutsche Forschungsgemeinschaft (DFG, German Research Fundation) under Germany’s Excellence Strategy EXC 2181/1 - 390900948 (the Heidelberg STRUCTURES Excellence Cluster).


If you use Norse in your work, please cite it as follows:

  author       = {Pehle, Christian and
                  Pedersen, Jens Egholm},
  title        = {{Norse -  A deep learning library for spiking 
                   neural networks}},
  month        = jan,
  year         = 2021,
  note         = {Documentation: https://norse.ai/docs/},
  publisher    = {Zenodo},
  version      = {0.0.6},
  doi          = {10.5281/zenodo.4422025},
  url          = {https://doi.org/10.5281/zenodo.4422025}

Norse is actively applied and cited in the literature. We are keeping track of the papers cited by Norse in our documentation.


LGPLv3. See LICENSE for license details.