Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/jqhoogland/randnn
Library for sampling random neural networks.
https://github.com/jqhoogland/randnn
Last synced: 10 days ago
JSON representation
Library for sampling random neural networks.
- Host: GitHub
- URL: https://github.com/jqhoogland/randnn
- Owner: jqhoogland
- License: mit
- Created: 2020-09-11T06:51:49.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2021-03-09T11:47:30.000Z (almost 4 years ago)
- Last Synced: 2024-11-19T18:55:05.889Z (2 months ago)
- Language: Python
- Homepage:
- Size: 6.66 MB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Random neural networks
A library for sampling random neural networks.
### The model
```math
\tau \frac{dh_i}{dt}=-h_i+\sum_{j=1}^N J_{ij}\phi(h_j),
```
where:
- $`h_i(t)`$ is the state of neuron $i$ at time $t$.
- $`J_{ij}`$ is the weight of the connection from neuron $j$ to neuron $i$.
- It is sampled from $\mathcal{N}(0, g^2/N)$, where $g$ is our coupling strength.
- $`J_{ii}=0`$
- $`\phi`$ is the activation function ($`=\tanh`$)