PyVista and QT

Demonstrate how to use PyVista to create standalone applications using pyinstaller and the Qt framework.

Tip

This section of the tutorial was adopted from the pyvistaqt documentation and the Extras chapters of the PyVista documentation.

Using PyVista with QT

PyVista has an interface for placing plots in QT (see QT) in the pyvistaqt library that allows you to combine QT with VTK. This simplifies adding meshes, updating, and controlling them when using Qt.

Installation using pip is:

pip install pyvistaqt

To install this package with conda run:

conda install -c conda-forge pyvistaqt

Once installed, use the pyvistaqt.BackgroundPlotter like any PyVista plotter.

Brief Example

Create an instance of the pyvistaqt.BackgroundPlotter and plot a sphere.

import pyvista as pv
from pyvistaqt import BackgroundPlotter

sphere = pv.Sphere()

plotter = BackgroundPlotter()
plotter.add_mesh(sphere)

Sample PyQt5 pyvista QtInteractor

Background Plotting

Normal PyVista plotting windows exhibit blocking behavior, but it is possible to plot in the background and update the plotter in real-time using pyvistaqt.BackgroundPlotter. This requires pyvistaqt, but otherwise appears and functions like a normal PyVista pyvista.Plotter instance. For example:

import pyvista as pv
from pyvistaqt import BackgroundPlotter

sphere = pv.Sphere()

plotter = BackgroundPlotter()
plotter.add_mesh(sphere)

# can now operate on the sphere and have it updated in the background
sphere.points *= 0.5

Multiple Plotters

The following example shows how to use an interface with multiple plotters. Each plotter can be selected and functions like a normal PyVista pyvista.Plotter instance:

import pyvista as pv
from pyvistaqt import MultiPlotter

mp = MultiPlotter(nrows=2, ncols=2)
mp[0, 0].add_mesh(pv.Sphere())
mp[0, 1].add_mesh(pv.Cylinder())
mp[1, 0].add_mesh(pv.Cube())
mp[1, 1].add_mesh(pv.Cone())

Freezing PyVista with pyinstaller

You can make some fantastic standalone programs with pyinstaller and PyVista, and you can even make a graphical user interface incorporating PyQt5 or pyside2. Depending on your version of VTK, this requires some extra steps to setup.

When running VTK v9, you need to add several additional hiddenimports. For clarity and completeness, create a spec file (we’ll name it pyvista.spec) following the directions given at Using Spec Files. Modify the Analysis and add the following hidden imports:

main_py = os.path.join(some_path, 'main.py')
a = Analysis([main_py],
             pathex=[],
             binaries=[],
             hiddenimports=['vtkmodules',
                            'vtkmodules.all',
                            'vtkmodules.qt.QVTKRenderWindowInteractor',
                            'vtkmodules.util',
                            'vtkmodules.util.numpy_support',
                            'vtkmodules.numpy_interface.dataset_adapter',
                           ],

From there, you can freeze an application using PyVista and create a standalone application.

Examples

VesselVioMNEfemorph

VesselVio is open-source application for the analysis and visualization of segmented vasculature datasets.