https://github.com/ornl/superneuro
Neuromorphic simulator with two modes of computation
https://github.com/ornl/superneuro
Last synced: about 2 months ago
JSON representation
Neuromorphic simulator with two modes of computation
- Host: GitHub
- URL: https://github.com/ornl/superneuro
- Owner: ORNL
- Created: 2023-03-27T15:29:06.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2023-10-25T19:47:07.000Z (over 1 year ago)
- Last Synced: 2025-03-24T20:23:05.840Z (2 months ago)
- Language: Jupyter Notebook
- Size: 6.02 MB
- Stars: 7
- Watchers: 9
- Forks: 1
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# SuperNeuro
Neuromorphic simulator with two modes of computation:
1. Agent Based Mode (ABM) - SuperNeuroABM
2. Matrix Computation - SuperNeuroMAT
## Installation: SuperNeuroABM
**Requirements:**
Python version >=3.8
**Optional requirements if running on GPU:**
CUDA enabled GPU with compute compatibility 6.0+
CUDA drivers
`pip install superneuroabm`
Or if the above doesn't work, you can try this:
`pip install git+https://github.com/ORNL/superneuroabm`
### Usage:
`from superneuroabm.model import NeuromorphicModel`
SuperNeuroABM allows one to run simulations on GPUs. It can be accessed by instantiating the model as
`NeuromorphicModel(use_cuda = True)`
## Intstallation: SuperNeuroMAT
Requirements: Python version >=3.8
`pip install superneuromat`
Or if the above doesn't work try the following from the github repo:
`pip install git+https://github.com/ORNL/superneuromat.git`
### Usage:
`from superneuromat.neuromorphicmodel import NeuromorphicModel`
In each of the modes, the neuromorphicmodel allows the user to create neurons and synapses as per the application or algorithm requirements.
## Tutorials:
Jupyter notebooks with tutorials for each of the two modes of SuperNeuro are located in the 'Tutorials' directory.
## Publications:
Arxiv Preprint available at:
[Date. P., Gunaratne, C., Kulkarni, S., Patton, R., Coletti, M., & Potok, T. (2023). SuperNeuro: A Fast and Scalable Simulator for Neuromorphic Computing. arXiv preprint arXiv:2305.02510.
](https://arxiv.org/abs/2305.02510)