This is a gaussian emulator and Baysiean analysis with PyMC

Installation with anaconda (Should work on both linux and windows machine) (Update 09/30/2023: Use mamba instead! Conda takes forever to solve environment!)

1. Install anaconda/miniconda for python3 by following instruction on https://docs.conda.io/en/latest/miniconda.html.
2. Update conda and add installation channel:

    $> conda update --all
    $> conda config --append channels conda-forge

3. Install mamba and activate environment.

     $> conda install -c conda-forge mamba
     $> conda activate

4. Go to /path/to/Bayesian-package and create virtual environment by running the following command:

    $> mamba env create --name <env_name> --file=environmentCrossPlatform.yml

5. Activate the environment:

    $> conda activate <env_name>

6. Start the program with

   $> python -m GUI.GUIController.GUIControllerMP <number-of-cores>

IF the script refuses to use more than one core despite > 1, try to restart the terminal after step 4. Sometimes the installation of mamba messes up the CPU affinity. Restarting after installation usually fixes it.

If not, use

    $> OPENBLAS_MAIN_FREE=1 python -m GUI.GUIController.GUIControllerMP <number-of-cores>

Running GUI on HPCC

This section only talks about running GUI in HPCC. Some changes to the installation procedure are needed because some version of MPI are not compatible with HPCC.

TBD

Using GUI

TBD