Download and manipulate neuromorphic datasets fast and easily!
Tonic provides publicly available event-based vision and audio datasets and event transformations. The package is fully compatible with PyTorch Vision / Audio to give you the flexibility that you need.
Getting started#
Install Tonic via pypi or anaconda.
Run a first example with this neuromorphic version of the MNIST dataset.
Tutorials#
If you want you can run them yourself in your browser using Binder.
Load images alongside events and apply augmentations.
Learn how to load data fast using disk-caching.
Batching when using events is straightforward.
Slice your dataset into smaller chunks if you need to.
How to work with larger datasets that output multiple data for heavy-duty processing.
How Tos#
Check out these scripts if you run into a specific problem.
API reference#
List of neuromorphic datasets. Vision and audio datasets.
List of event transformations. Event transforms and representations.
{doc}Supported file parsers. For the various file formats out there.
Reading material#
Introduction to neuromorphic cameras if you’ve never worked with events/spikes.
Short intro to spiking neural networks and how they work with events.
Links to external spiking neural network simulators to train your network.
Read about design decisions we made in Tonic.
Getting involved#
Contribution guidelines. Please read this before opening a pull request.
Communication channels to get in touch.
About#
About Tonic. How the project came to life.
Release notes. Version changes.