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

argsort

def argsort[*, ascending: Bool = True, target: StringSlice[ImmStaticOrigin] = StringSlice("cpu")](output: TileTensor[Storage=output.Storage, linear_idx_type=output.linear_idx_type], input: TileTensor[Storage=input.Storage, address_space=input.address_space, linear_idx_type=input.linear_idx_type], ctx: DeviceContext)

Performs argsort on input buffer, storing indices in output buffer.

Parameters:

• ​ascending (Bool): Sort direction (True for ascending, False for descending).
• ​target (StringSlice[ImmStaticOrigin]): Target device ("cpu" or "gpu").

Args:

• ​output (TileTensor[Storage=output.Storage, linear_idx_type=output.linear_idx_type]): Buffer to store sorted indices.
• ​input (TileTensor[Storage=input.Storage, address_space=input.address_space, linear_idx_type=input.linear_idx_type]): Buffer containing values to sort.
• ​ctx (DeviceContext): Device context for execution.

def argsort[ascending: Bool = True](output: TileTensor[Storage=output.Storage, linear_idx_type=output.linear_idx_type], input: TileTensor[Storage=input.Storage, address_space=input.address_space, linear_idx_type=input.linear_idx_type])

CPU-only version of argsort.

Parameters:

• ​ascending (Bool): Sort direction (True for ascending, False for descending).

Args:

• ​output (TileTensor[Storage=output.Storage, linear_idx_type=output.linear_idx_type]): Buffer to store sorted indices.
• ​input (TileTensor[Storage=input.Storage, address_space=input.address_space, linear_idx_type=input.linear_idx_type]): Buffer containing values to sort.