xlnstorch.get_implementation#
- xlnstorch.get_implementation(torch_function, impl_key)#
Get the implementation tuple for a given torch function and implementation key.
- Parameters:
torch_function (Callable) – The torch function for which to get the implementation.
impl_key (str) – The key identifying the specific implementation.
- Returns:
An implementation object containing the function that implements the LNS operation, and the internal computation function.
- Return type:
Implementation
- Raises:
ValueError – If no implementations are registered for the given torch function. If the specified implementation key is not registered for the torch function.