fields.tensors.DenseTensorField.__init__

DenseTensorField.__init__(grid: GridBase, component: FieldComponent, signature: Sequence[Literal[1, -1]] | None = None)

Initialize a dense tensor field over a structured coordinate grid.

Parameters:

• grid (GridBase) – The structured grid over which the field is defined. This determines the spatial axes, metric, and chunking behavior.

• component (FieldComponent) –

  The single component storing the field data. The component must have trailing element dimensions that match the expected tensor shape: one dimension per tensor index, each of size equal to grid.ndim.

  For example, a rank-2 tensor field in 3D requires component.element_shape = (3, 3).

• signature (list of {+1,-1}, optional) –

  Tensor signature indicating the variance of each index:

  • +1 → contravariant index (superscript)

  • -1 → covariant index (subscript)

  If not specified, a default signature of all +1 (fully contravariant) is assumed.

Raises:

ValueError – If the component’s element shape is not consistent with the tensor rank or the grid dimensionality, or if the signature is invalid.