Data repository for "Accelerating Spiking Neural Networks on CPUs via Cache-aware Splitting"

DOI

This data repository contains the experiment data and code for the paper "Accelerating Spiking Neural Networks on CPUs via Cache-aware Splitting".

The code.zip file contains the git repositories of the code that was used in the experiments:

fast-arrays library: executes vectorized calculations with AVX-512F operations
mlflow-rs library: client for the MLflow experiment tracking tool to store metrics and artifacts of the experiments
snn-rs project: SNN implementation as well as experiment execution code

The experiment_tracking.zip file contains the experiment results:

hyperparameters
the exact code that was executed (git commit hash + git patch)
collected metrics
trained SNN models
class assignments
training log
Identifier
DOI https://doi.org/10.18419/DARUS-5267
Metadata Access https://darus.uni-stuttgart.de/oai?verb=GetRecord&metadataPrefix=oai_datacite&identifier=doi:10.18419/DARUS-5267
Provenance
Creator Wundrack, Philipp ORCID logo; Barzen, Johanna ORCID logo; Bechtold, Marvin ORCID logo
Publisher DaRUS
Contributor Wundrack, Philipp
Publication Year 2025
Rights info:eu-repo/semantics/openAccess
OpenAccess true
Contact Wundrack, Philipp (University of Stuttgart)
Representation
Resource Type Dataset
Format application/octet-stream
Size 534369; 289505973
Version 1.0
Discipline Other