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

Mojo function

scatter_elements_shape

def scatter_elements_shape[input_type: DType, indices_type: DType](input: TileTensor[input_type, Storage=input.Storage, address_space=input.address_space, linear_idx_type=input.linear_idx_type, element_size=input.element_size], updates: TileTensor[input_type, Storage=updates.Storage, address_space=updates.address_space, linear_idx_type=updates.linear_idx_type, element_size=updates.element_size], indices: TileTensor[indices_type, Storage=indices.Storage, address_space=indices.address_space, linear_idx_type=indices.linear_idx_type, element_size=indices.element_size], axis: Int) -> IndexList[input.LayoutType.rank]

Compute the output shape of a scatter_elements operation, and assert the inputs are compatible.

Parameters:

input_type (DType): Type of the input tensor.
indices_type (DType): Type of the indices tensor.

Args:

input (TileTensor[input_type, Storage=input.Storage, address_space=input.address_space, linear_idx_type=input.linear_idx_type, element_size=input.element_size]): The input tensor.
updates (TileTensor[input_type, Storage=updates.Storage, address_space=updates.address_space, linear_idx_type=updates.linear_idx_type, element_size=updates.element_size]): The input tensor.
indices (TileTensor[indices_type, Storage=indices.Storage, address_space=indices.address_space, linear_idx_type=indices.linear_idx_type, element_size=indices.element_size]): The indices tensor.
axis (Int): The axis.

Returns:

IndexList[input.LayoutType.rank]: The output shape.