Using PyVista & Trame with Panel

Visit the demo at pyvista-panel-example.

This is an example of using PyVista & Trame with Panel. It's designed to be able to be deployed on the cloud (e.g. GCP) as a containerized application that only relies on a single port being exposed .

Normally, Trame requires you to open up a second port in addition to the default to the Panel port, but here we use jupyter-server-proxy to proxy the traffic to an endpoint of our choosing, in this case, /proxy. This way, we can get the benefits of using the Trame with PyVista and being able to deploy on the cloud.

While, Panel already has a VTK extension, it uses an old serializer, meaning you'll be missing out in all the latest changes from Trame.

Running the Application Locally

You can run the application locally with and without Docker.

With Docker Compose

make build
make serve-docker

Visit http://localhost:8080

Without Docker

pip install -r app/requirements.txt
pip install setuptools pyvista
make serve

Note the two extra requirements since they're normally included in the docker image.

Visit http://localhost:8080

Development

This repository uses pip-compile to generate a requirements.txt file from requirements.in, locking down your requirements given a set of dependencies. This is used to ensure that the version of all of your requirements (e.g. dependencies of dependencies) do not change when the docker image is built. Run this with:

pip install pip-tools
make compile-requirements

Deployment on the Cloud

This app is designed to be hosted as a containerized application on the cloud.

This application can be deployed on major cloud providers by either generating and deploying the docker image, or in the case of GCP, simply running gcloud run deploy and deploying directly from source, as can be useful in development.

Here's an example:

gcloud run deploy pyvista-demo --source=/home/user/source/panel-example --region=us-central1 --project=reliable-house-439292-m4