Architecture Design Space Graph Core
Project description
Architecture Design Space Graph Core
GitHub Repository | Documentation
The Architecture Design Space Graph (ADSG) allows you to model design spaces using a directed graph that contains three types of architectural choices:
- Selection choices (see example below): selecting among mutually-exclusive options, used for selecting which nodes are part of an architecture instance
- Connection choices: connecting one or more source nodes to one or more target nodes, subject to connection constraints and optional node existence (due to selection choices)
- Additional design variables: continuous or discrete, subject to optional existence (due to selection choices)
Modeling a design space like this allows you to:
- Model hierarchical relationships between choices, for example only activating a choice when another choice has some option selected, or restricting the available options for choices based on higher-up choices
- Formulate the design space as an optimization problem that can be solved using numerical optimization algorithms
- Generate architecture instances for a given design vector, automatically correct incorrect design variables, and get information about which design variables were active
- Implement an evaluation function (architecture instance --> metrics) and run the optimization problem
Installation
First, create a conda environment (skip if you already have one):
conda create --name adsg python=3.10
conda activate adsg
Then install the package:
conda install numpy scipy~=1.9
pip install adsg-core
Optionally also install optimization algorithms (SBArchOpt):
pip install adsg-core[opt]
If you want to interact with the ADSG from a Jupyter notebook:
pip install adsg-core[nb]
jupyter notebook
Documentation
Refer to the documentation for more background on the ADSG and how to implement architecture optimization problems.
Examples
An example ADSG with two selection choices:
An example ADSG with a connection choice:
The ADSG of the Apollo problem:
The ADSG of the GNC problem:
Citing
If you use the ADSG in your work, please cite it:
Bussemaker, J.H., Ciampa, P.D., & Nagel, B. (2020). System architecture design space exploration: An approach to modeling and optimization. In AIAA Aviation 2020 Forum (p. 3172). DOI: 10.2514/6.2020-3172
Contributing
The project is coordinated by: Jasper Bussemaker (jasper.bussemaker at dlr.de)
If you find a bug or have a feature request, please file an issue using the Github issue tracker. If you require support for using ADSG Core or want to collaborate, feel free to contact me.
Contributions are appreciated too:
- Fork the repository
- Add your contributions to the fork
- Update/add documentation
- Add tests and make sure they pass (tests are run using
pytest
)
- Read and sign the Contributor License Agreement (CLA) , and send it to the project coordinator
- Issue a pull request into the
dev
branch
NOTE: Do NOT directly contribute to the adsg_core.optimization.assign_enc
and .sel_choice_enc
modules!
Their development happens in separate repositories:
Use update_enc_repos.py
to update the code in this repository.
Adding Documentation
pip install -r requirements-docs.txt
mkdocs serve
Refer to mkdocs and mkdocstrings documentation for more information.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for adsg_core-1.0.2-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | a0c61e0b4feac79e4adf18cfe1465a058be859dce1c60154a63d92283914d0dc |
|
MD5 | ff74c8ea8883cd0c6885ce20b0517445 |
|
BLAKE2b-256 | d8b1053057f951e2d1683c7494c0e8992c721c94f7f247256f1f809dfd4342f0 |