enweg/quickbnnr: Quick way to implement Bayesian Neural Networks in R and estimating these using Julia's Turing
Allows for a quick and flexible specification of Bayesian Neural Networks. These networks can be specified as easiliy as writing Chain(DenseBNN(1, 10, "sigmoid"), DenseBNN(10, 1)) and are then transformed to Turing models. Turing is like Stan but completely implemented in Julia. The models are then estimated using the NUTS sampler and bayesplot can be used to investigate the chains.
Getting started
Browse package contents
Package details |
|
|---|---|
| Author | Enrico Wegner |
| Maintainer | Enrico Wegner <e.wegner@student.maastrichtuniversity.nl> |
| License | MIT + file LICENSE |
| Version | 0.1.0 |
| Package repository | View on GitHub |
| Installation |
Install the latest version of this package by entering the following in R:
|
