festim-gui

GUI for FESTIM

License

This library is OpenSource and follow the MIT License

Installation

Install the application/library

pip install festim-gui

Run the application

festim-gui

Simulation runs

The Run page uses FESTIM_GUI_PYTHON to execute generated scripts, falling back to the interpreter running festim-gui. The Docker image configures this variable automatically. Outside Docker, install festim in the active environment or set the variable to an interpreter that provides it.

Simulation working directories are created under the system temporary directory (e.g. /tmp).

Each run gets its own temporary directory containing the generated script, a run.log file, and any simulation outputs written by the script.

Development setup

We recommend using uv for setting up and managing a virtual environment for your development.

# Create venv and install all dependencies
uv sync --all-extras --dev

# Activate environment
source .venv/bin/activate

# Install commit analysis
pre-commit install
pre-commit install --hook-type commit-msg

For running tests and checks, you can run nox.

# run all
nox

# lint
nox -s lint

# tests
nox -s tests

Docker

Build the Docker image from the repository root:

docker build -t festim-gui .

Run the image and expose it on port 8080:

docker run -it --rm -p 8080:80 festim-gui

For development, you can mount the source code into the container:

docker run -it --rm -p 8080:80 -v .:/local-app festim-gui

Then open http://localhost:8080/ in your browser.

The Docker image includes festim, so Run page executions work inside the container without any extra local FESTIM installation.

Docker setup files are located under setup/.

GPU support

The post-processing view can render either on the client or on the server:

• Local rendering (default) -- geometry is sent to the browser and rendered in browser. No GPU is needed on the server.
• Remote rendering (GPU) -- ParaView renders on the server and streams images to the browser. This requires a GPU available to the process, and is enabled with the --gpu flag.

The Docker image exposes both modes as separate apps, declared in setup/apps.yml:

• http://localhost:8080/index.html -- default app, local rendering.
• http://localhost:8080/gpu.html -- same app launched with --gpu, remote rendering.

Running the container with GPUs

Install the NVIDIA Container Toolkit on the host, then pass the GPUs to the container with --gpus:

docker run -it --rm --gpus all -p 8080:80 festim-gui

Then open http://localhost:8080/gpu.html in your browser.

Professional Support

• Training: Learn how to confidently use trame from the expert developers at Kitware.
• Support: Our experts can assist your team as you build your web application and establish in-house expertise.
• Custom Development: Leverage Kitware’s 25+ years of experience to quickly build your web application.