Battling with all the different file formats of publicly available neuromorphic datasets? No more! Tonic is a tool to facilitate the download, manipulation and loading of event-based/spike-based data. Have a look at the list of supported datasets and transformations! It's somewhat modeled after PyTorch Vision for an intuitive interface, so that you spend less time worrying about how to read files and more time on things that matter.
pip install tonicIf you prefer conda, please check out the forge repository.
Have a look at our introduction page to see how some of the moving parts work. There are some more short examples available here.
If you're looking for a minimal example to run, this is it!
import tonic
import tonic.transforms as transforms
transform = transforms.Compose([transforms.Denoise(filter_time=10000),
transforms.TimeJitter(std=10),])
testset = tonic.datasets.NMNIST(save_to='./data',
train=False,
transform=transform)
from torch.utils.data import DataLoader
testloader = DataLoader(testset, shuffle=True)
events, target = next(iter(testloader))Have a question about how something works? Ideas for improvement? Feature request? Please get in touch here on GitHub via the Discussions page!
You can find the full documentation on Tonic on this site.
If you find this package helpful, please use the following citation:
@software{lenz_gregor_2021_5079802,
author = {Lenz, Gregor and
Chaney, Kenneth and
Shrestha, Sumit Bam and
Oubari, Omar and
Picaud, Serge and
Zarrella, Guido},
title = {Tonic: event-based datasets and transformations.},
month = jul,
year = 2021,
note = {{Documentation available under
https://tonic.readthedocs.io}},
publisher = {Zenodo},
version = {0.4.0},
doi = {10.5281/zenodo.5079802},
url = {https://doi.org/10.5281/zenodo.5079802}
}