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

top_k_shape

def top_k_shape(input: T, k: Scalar, axis: Scalar, sorted: Scalar[DType.bool]) -> IndexList[T.rank]

Computes the output shape for the mo.top_k graph op.

Args:

• ​input (T): Input tensor to select the top-k values from.
• ​k (Scalar): Number of largest elements to select along axis.
• ​axis (Scalar): Dimension along which to select the top-k elements.
• ​sorted (Scalar[DType.bool]): Whether to sort the selected elements in descending order.

Returns:

IndexList[T.rank]: The output shape after selecting the top-k values along axis.