Skip to main content

A framework for performing forward and derivative analyses for systems following directed acylic graph structure.

Project description

Build and deploy webpage Lint Run Unittests

Flume

Flume is an open-source framework for multidisciplinary analysis and adjoint evaluations based on directed acyclic graphs. It is intended to be a lightweight framework that affords users with a degree of flexibility in implementing their own analysis and optimization procedures while providing a common underlying structure. Flume also has a set of interfaces between some common optimizers, which allow the user to easily set up an optimization problem that corresponds to their specific needs.

Getting Started

For new users, it is recommended that you read the Flume overview notebook, which outlines in detail the critical aspects of the framework in context of an example. This discusses the nomenclature that is used within the framework and illustrates how to construct State, Analysis, and System objects for use with a Flume interface. Also, for a review of other useful methods that are included for Analysis classes, check out the methods demonstration notebook. The examples gallery also includes several different types of problems, which showcase the construction of many types of Analysis classes, the assembly of a System, and how to interface with an optimizer interface.

Installation

To install Flume, the only necessary dependencies are a working version of Python that is at least version 3.10 and Graphviz. For Graphviz, check out the installation instructions located here depending on your machine type. It is generally encouraged that you use a virtual environment. If you do not have a virtual environment already, navigate to the directory where you want to use Flume and execute the following.

python -m venv venv
source venv/bin/activate

Then, with the virtual environment activated, simply install the package from PyPi with

pip install pip install flume-smdo==1.0.0

This will install the base classes, optimizer interfaces, and utilities into a Python package named "flume". For those who want to make changes to the code or access the unit tests and examples, you can also clone the repository and then install in editable mode.

cd {your_chosen_git_directory}
git clone https://github.com/smdogroup/flume.git
cd flume
python -m venv venv
source venv/bin/activate
pip install -e .

After following either of these processes, you should be able to access the base classes and get started with developing your own scripts within the framework! It should be noted that ParOpt is a separate package that is not included with Flume during the build process. For those who want to utilize this interface, we refer you to the ParOpt installation instructions.

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

flume_smdo-1.0.1.tar.gz (20.5 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

flume_smdo-1.0.1-py3-none-any.whl (24.2 MB view details)

Uploaded Python 3

File details

Details for the file flume_smdo-1.0.1.tar.gz.

File metadata

  • Download URL: flume_smdo-1.0.1.tar.gz
  • Upload date:
  • Size: 20.5 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.10.8

File hashes

Hashes for flume_smdo-1.0.1.tar.gz
Algorithm Hash digest
SHA256 0859a83d6239d3e92e6cb353e226fb79f275199d5a05ad97be7a14751abe0251
MD5 c0520cc6abb9a99b61bfded4c59a0975
BLAKE2b-256 fe5f025e409e670097ba496fee657ca97e23fa09ac91d15db1a8eca4edb4aa98

See more details on using hashes here.

File details

Details for the file flume_smdo-1.0.1-py3-none-any.whl.

File metadata

  • Download URL: flume_smdo-1.0.1-py3-none-any.whl
  • Upload date:
  • Size: 24.2 MB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.10.8

File hashes

Hashes for flume_smdo-1.0.1-py3-none-any.whl
Algorithm Hash digest
SHA256 7fbc16ba2ec997d7ee1aa80180a6d5b9c2a99945e15a54f870f4510b96f83114
MD5 2a5a7dc0858ffaea9bba7d54354ba5d5
BLAKE2b-256 8f63a66e0c97f6cb7d9602f1635022596665b9a6a30d1b98be51638a2d657f52

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page