Neuronal Gaussian Process Regression

This repository is the official implementation of Neuronal Gaussian Process Regression.

Requirements

To install requirements (using conda) and download the datasets, execute:

conda env create -f environment.yml
conda activate neuronalGPR
python setup.py

Training

Pre-trained models are included in the results directory of this repository. To nevertheless re-train the models on the UCI datasets, run these commands:

python runUCI.py <UCI Dataset directory>
python PBP.py <UCI Dataset directory> <number of hidden layers>
python Dropout.py <UCI Dataset directory> <number of hidden layers>

Evaluation

To reproduce the figures, run

python <name_of_fig_script.py>

To reproduce the tables, run

python table1.py; python table2.py; 

Pre-trained Models

The pre-trained models are included in the results directory of this repository.