For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Mojo function

crd2idx

def crd2idx(crd: IntTuple, shape: IntTuple) -> Int

Map a logical coordinate to a linear index.

This function converts a multi-dimensional coordinate to a linear index based on the shape. It uses default strides computed from the shape.

Args:

crd (IntTuple): The coordinate tuple to convert.
shape (IntTuple): The shape of the tensor/array.

Returns:

Int: The linear index corresponding to the coordinate.

def crd2idx(crd: IntTuple, shape: IntTuple, _stride: IntTuple) -> Int

Map a logical coordinate to a linear index with custom strides.

This function converts a multi-dimensional coordinate to a linear index based on the shape and stride information. If no stride is provided, it computes default strides from the shape.

The function handles various input combinations:

Tuple coordinates with tuple shapes and strides
Single integer coordinate with tuple shapes and strides
Single integer coordinate with single integer shape and stride

Aborts:

- If coordinate and shape dimensions don't match.
- If shape and stride dimensions don't match.
- If input type combinations are invalid.

Args:

crd (IntTuple): The coordinate(s) to convert, can be a single value or a tuple of coordinates.
shape (IntTuple): The shape of the tensor/array, can be a single value or a tuple of dimensions.
_stride (IntTuple): Optional custom strides, defaults to row-major strides if not provided.

Returns:

Int: The linear index corresponding to the coordinate.