psimpy 1.0.1

Predictive and probabilistic simulation tools.

Explore the docs »

View Publication · View Example · Report Bug · Request Feature

Table of Contents

• Description
• Prerequisites
• Installation
  • Installation in a Conda Environment (Recommended)
• Documentation
• Usage
• Cite as
  • Corresponding publications
• Research Studies Using Psimpy
• License

Description

PSimPy (Predictive and probabilistic simulation with Python) implements a Gaussian process emulation-based framework that enables systematic and efficient investigation of uncertainties associated with physics-based models (i.e. simulators).

Prerequisites

Before installing and using PSimPy, please ensure that you have the following prerequisites:
(Please note that we will cover number 1 to 3 in our recommended installation method: Installation in a Conda Environment.)

1. Python 3.9 or later:
   Make sure you have Python installed on your system. You can download the latest version of Python from the official website: Python Downloads
2. R Installed and Added to the PATH Environment Variable:
   • Install R from the official R Project website.
   • Add R to your system's PATH environment variable. This step is crucial for enabling communication between Python and R.
3. (Optional) RobustGaSP - R package:
   The emulator module, robustgasp.py, relies on the R package RobustGaSP. This has also been initegrated with other PSimPy modules, such as active_learning.py. In order to utilize these modules, make sure to install the R package RobustGaSP first.
4. (Optional) r.avaflow - Mass Flow Simulation Tool:
   PSimPy includes a simulator module, ravaflow3G.py, that interfaces with the open source software r.avaflow 3G. If you intend to use this module, please refer to the official documentation of r.avaflow 3G to for installation guide.

Installation

PSimPy can be installed using pip.

$ pip install psimpy

This command will install the package along with its dependencies.

Installation in a Conda Environment (Recommended)

We recommond you to install PSimPy in a virtual environment such as a conda environment. In this section, we will ceate a conda environment with prerequisites (number 1 to 3), and install PSimPy in this environment. You may want to first install Anaconda or Miniconda if you haven't. The steps afterwards are as follows:

1. Create a conda environment with Python, R, and RobustGaSP, and activate the environment:

   conda create --name your_env_name python r-base conda-forge::r-robustgasp
   conda activate your_env_name
   

2. Install PSimPy using pip in your conda environment:

   pip install psimpy
   

Now you should have PSimPy and its dependencies successfully installed in your conda environment. You can use it in the Python terminal or in your Python IDE.

Quick Note on R_HOME in Conda Environments:

If you're running PSimPy in a conda environment without a predefined R_HOME variable, we automatically set it to the default R installation path of the active conda environment. This ensures PSimPy works smoothly with R without needing manual setup. If you prefer setting R_HOME yourself, please define it before starting PSimPy to use a custom R environment.

Documentation

Detailed documentation of PSimPy is hosted at https://mbd.pages.rwth-aachen.de/psimpy/, including the API and theory (or reference) of each module.

Usage

Usage examples are provided by the Example Gallery.

Cite as

@misc{psimpy,
  author = {Hu Zhao},
  title = {PSimPy : Predictive and probabilistic simulation with Python},
  year = {2022},
  howpublished = {\url{https://git.rwth-aachen.de/mbd/psimpy}},
}

Corresponding publications

Hu Zhao, Anil Yildiz, Nazanin Bagherinejad, Julia Kowalski, PSimPy: GP emulation-based sensitivity analysis, uncertainty quantification and calibration of landslide simulators, 14th International Conference on Applications of Statistics and Probability in Civil Engineering (ICASP14), Dublin, Ireland, 2023. http://hdl.handle.net/2262/103542

@InProceedings{zhao_et_al_2023,
  title       = {PSimPy: GP emulation-based sensitivity analysis, uncertainty quantification and calibration of landslide simulators},
  booktitle   = {14th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP14},
  author      = {Zhao, Hu and Yildiz, Anil and Bagherinejad, Nazanin and Kowalski, Julia},
  year        = {2023},
  address     = {{Dublin, Ireland}},
  note = {Available at \url{http://hdl.handle.net/2262/103542}},
}

Research Studies Using Psimpy

• Tillmann, S., Behr, M., & Elgeti, S. (2024). Using Bayesian optimization for warpage compensation in injection molding. Materialwissenschaft und Werkstofftechnik, 55(1), 13-20.
• Correa, A. (2024, July 10). Seamless Reproducibility of Complex Simulation Workflows [Conference presentation]. JuliaCon 2024, Eindhoven, Netherlands. https://pretalx.com/juliacon2024/talk/FJRZL7/
• Kumar, V. M. and Kowalski, J.: A unified Bayesian model selection workflow for geophysical free-surface flow, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18847, https://doi.org/10.5194/egusphere-egu24-18847, 2024.

License

PSimPy was created by Hu Zhao at the Chair of Methods for Model-based Development in Computational Engineering (RWTH Aachen University, Germany). It is licensed under the terms of the MIT license.