tiledbsoma.DenseNDArray.read

DenseNDArray.read(coords: Sequence[None | int | Slice[int]] = (), *, result_order: ResultOrder | Literal['auto', 'row-major', 'column-major'] = ResultOrder.ROW_MAJOR, partitions: ReadPartitions | None = None, platform_config: Dict[str, Mapping[str, Any]] | object | None = None) → Tensor

Reads a user-defined dense slice of the array and return as an Arrow Tensor.

Coordinates must specify a contiguous subarray, and the number of coordinates must be less than or equal to the number of dimensions. For example, if the array is 10 by 20, then some acceptable values of coords include (3, 4), (slice(5, 10),), and (slice(5, 10), slice(6, 12)). Slice indices are doubly inclusive.

Parameters:

  • coords – The coordinates for slicing the array.

  • result_order – Order of read results. This can be one of ‘row-major’ (default) or ‘column-major’

  • partitions – An optional ReadPartitions hint to indicate how results should be organized.

Raises:

SOMAError – If the object is not open for reading.

Lifecycle

Maturing.