Reinforcement Learning based Shape Optimization (ReLeSO)

Releso is a Library/Framework for Reinforcement Learning based Shape Optimization. Please look into the Documentation for information on how it works. The instruction on how the documentation can be built is given below as well as the instruction on how the package can be installed.
Alternatively, it can be installed from pip via pip install releso.

Documentation generation

Install and usage instructions are provided in the documentation of the project. The documentation can be built with the use of sphinx which is a python tool to generate documentation.

The sphinx packages can either be installed in the project python environment or in a separate environment. If it does not matter in which python environment sphinx is installed ignore the first two lines.

The following command line calls create a conda environment with all necessary dependencies for building the documentation.

(base) $ conda create -n sphinx python=3.9
(base) $ conda activate sphinx
(sphinx) $ pip install ".[docs]"

The documentation is built by executing the following command inside the folder docs/. After executing the command the documentation should be available inside the folder docs/build/html/

(sphinx) $ make html

Installation

This section covers the installation process of the framework and its prerequisites. The first thing to note is that with version 0.1.0 the strict dependency on splinepy is not present anymore. But if the geometry is to be parameterized by a Spline and the method of Free Form Deformation is to be used to deform a mesh, splinepy is still necessary.

Prerequisites

To use ReLeSO the following packages have to be installed:

• pydantic<2
• stable-baselines3
• tensorboard
• hjson

The pydantic package currently needs to be on version 1.*, we welcome anyone wanting to update releso to the new pydantic version.

The packages can be installed via pip or conda with the following commands:

pip (activation of the venv should be done beforehand)

(.venv) $ pip install pydantic stable-baselines3 tensorboard hjson

conda

(base) $ conda create -n releso python=3.9 "pydantic<2" tensorboard
(base) $ conda activate releso
(releso) $ pip install stable-baselines3 hjson

If the spline-based shape optimization functionality is needed, the package splinepy is needed. Please visit splinepy on github for installation instructions.

Development

To develop the framework further the sphinx package should also be installed with the currently used sphinx html theme sphinx_rtd_theme. This can be done via:

(releso) $ pip install sphinx sphinx_rtd_theme

Framework

After installing all prerequisites the framework itself can be installed by running the command below in the main repository folder.

Non-development

(releso) $ pip install .

Development

(releso) $ pip install -e "dev."