This repository contains the implementation of the code for the paper "BayesSum".
To install the required packages, run the following command:
pip3 install -r requirements.txt- Synthetic Data To reproduce the results for the synthetic experiment Figure 1 and Figure 2(Left), run the following files:
- Poisson:
python3 poisson.py - Uniform:
python3 potts.py --figure uniform(Figure 1)python3 potts.py --figure slope(Table of convergence rate)python3 potts.py --figure dim_boxplot(Figure 2 Right),python3 potts.py --figure lambda_ablation(Figure 2 Left) - Potts model:
python3 unnormalised_potts.py - Mixed space:
python3 mixed_bq.py
-
Sales Data To reproduce the results for the synthetic experiment Figure 2(Right) and Figure 3, run the following command:
python3 cmp_salesdata.py
-
Synthetic Protein Sequence To reproduce the results for the synthetic experiment Figure 4, run the following command:
python3 protein_potts_mll_mix.py