pyrho.pgrid#

Python class for ND grid data volumetric data.

Module Contents#

Classes#

PGrid

Class representing of _periodic_ grid data.
class pyrho.pgrid.PGrid(grid_data: numpy.typing.NDArray, lattice: numpy.typing.NDArray)[source]#

Bases: monty.json.MSONable

Class representing of _periodic_ grid data.

Represents a periodic scalar field on a regular grid. The data is defined by the lattice vectors and the grid data. The grid points are implicitly defined by the lattice vectors and the grid shape.

_transform_data(sc_mat: numpy.typing.ArrayLike, grid_out: numpy.typing.ArrayLike, origin: numpy.typing.ArrayLike | None = None, up_sample: int = 1) → numpy.typing.NDArray[source]#

Apply a supercell transformation to the grid data.

This function assumes that the data is fixed in place and the transformation is applied to the lattice vectors.

Parameters:

  • sc_mat – Matrix transformation applied to the lattice vectors

  • grid_out – The dimensions of the output grid

  • origin – Origin of the new lattice in fractional coordinates of the input cell

  • up_sample – The factor to scale up the sampling of the grid data using Fourier interpolation

Returns:

The transformed data

Return type:

NDArray

__mul__(factor: float) → PGrid[source]#

Multiply the grid data by a factor.

Parameters:

factor – The factor to multiply the grid data by

Returns:

The new PGrid object

Return type:

PGrid

__truediv__(factor: float) → PGrid[source]#

Divide the grid data by a factor.

Parameters:

factor – The factor to divide the grid data by

Returns:

The new PGrid object

Return type:

PGrid

get_transformed(sc_mat: list[list[int]] | numpy.typing.NDArray, grid_out: list[int], origin: numpy.typing.NDArray | None = None, up_sample: int = 1) → PGrid[source]#

Get a new PGrid object for the new transformed data.

Parameters:

  • sc_mat – Matrix transformation applied to the lattice vectors

  • grid_out – The dimensions of the output grid

  • origin – Origin of the new lattice in fractional coordinates of the input cell

  • up_sample – The factor to scale up the sampling of the grid data using Fourier interpolation

Returns:

The transformed PGrid object

Return type:

PGrid

lossy_smooth_compression(grid_out: list, smear_std: float = 0.2) → numpy.typing.NDArray[source]#

Perform Fourier interpolation then Gaussian smoothing.

The smoothing makes sure that simple operation like max and min filters still give the same results.

Parameters:

  • grid_out – desired output grid of the compressed data.

  • smear_std – standard deviation of the Gaussian smoothing

Returns:

Smoothed array

Return type:

NDArray