Point Cell Scalars

This example demonstrates how to add point scalars for each individual cell to a dataset.

from __future__ import annotations

import numpy as np

from pyvista import examples

load the first 4 cells from the example UnstructuredGrid. Note how the number of points is less than 32 since all the points are joined in the center.

grid = examples.load_hexbeam().extract_cells(range(4))
grid

Header	Data Arrays
UnstructuredGrid Information N Cells 4 N Points 18 X Bounds 0.000e+00, 1.000e+00 Y Bounds 0.000e+00, 1.000e+00 Z Bounds 0.000e+00, 5.000e-01 N Arrays 5	Name Field Type N Comp Min Max sample_point_scalars Points int64 1 1.000e+00 2.860e+02 VTKorigID Points int64 1 0.000e+00 9.000e+01 vtkOriginalPointIds Points int64 1 0.000e+00 9.000e+01 sample_cell_scalars Cells int32 1 1.000e+00 4.000e+00 vtkOriginalCellIds Cells int64 1 0.000e+00 3.000e+00

Plot Point Scalars

At this point it’s possible to assign only point or cell scalars to this dataset. First, let’s just plot some simple point scalars.

grid.point_data['Point Data'] = range(grid.n_points)
grid.plot(scalars='Point Data')

Static Scene

Interactive Scene

Plot Cell Scalars

Next, let’s plot cell scalars. We’re simply assigning based on the cell index.

grid.cell_data['Cell Data'] = range(grid.n_cells)
grid.plot(scalars='Cell Data')

Static Scene

Interactive Scene

Splitting the Cells

If you wanted to assign data to each point of each cell and plot that, it’s simply not possible since these hexahedral cells all share the same points. To split up individual cells, separate them using pyvista.DataSetFilters.separate_cells().

With this filter the resulting pyvista.UnstructuredGrid now contains 32 points, or 8 for each cell. They are now fully separated with no shared points.

split_cells = grid.separate_cells()
split_cells

Header	Data Arrays
UnstructuredGrid Information N Cells 4 N Points 32 X Bounds 0.000e+00, 1.000e+00 Y Bounds 0.000e+00, 1.000e+00 Z Bounds 0.000e+00, 5.000e-01 N Arrays 7	Name Field Type N Comp Min Max sample_point_scalars Points int64 1 1.000e+00 2.860e+02 VTKorigID Points int64 1 0.000e+00 9.000e+01 vtkOriginalPointIds Points int64 1 0.000e+00 9.000e+01 Point Data Points int64 1 0.000e+00 1.700e+01 sample_cell_scalars Cells int32 1 1.000e+00 4.000e+00 vtkOriginalCellIds Cells int64 1 0.000e+00 3.000e+00 Cell Data Cells int64 1 0.000e+00 3.000e+00

Plot Point Cell Data

Now we can plot values for each point for each cell. This will still be assigned to the point data.

Here we use numpy.hstack() for clarity, but as long as the length of the data matches the number of points, you’ll be able to use this approach.

See how the plotted values appear continuous within a cell and discontinuous between cells. This matches how stresses and strains are calculated from finite element solutions.

split_cells.point_data['Point Cell Data'] = np.hstack(
    (
        np.linspace(0, 8, 8),  # cell 0
        np.linspace(0, 12, 8),  # cell 1
        np.linspace(0, 16, 8),  # cell 2
        np.linspace(0, 20, 8),  # cell 3
    ),
)
split_cells.plot(scalars='Point Cell Data')

Static Scene

Interactive Scene