Method ApplyLoraAdapter
ApplyLoraAdapter(string, float)
Preloads a Low-Rank Adaptation (LoRA) adapter from the specified file into the model,
preparing it for dynamic application at inference time. Upon preload, a new entry
appears in the model’s Adapters collection.
The adapter will only take effect when its scale
is greater than zero.
public void ApplyLoraAdapter(string path, float scale = 1)
Parameters
path
stringThe file system path to the LoRA adapter source file. Must not be null, empty, or whitespace.
scale
floatThe initial scale factor for the LoRA adapter. Must be zero or positive. A value of zero registers the adapter but keeps it inactive until the scale is raised, at which point it will be applied during inference.
Exceptions
- ArgumentNullException
Thrown if the
path
parameter is null, empty, or whitespace.- ArgumentOutOfRangeException
Thrown if the
scale
parameter is less than zero.- FileNotFoundException
Thrown if no file exists at the specified
path
.- InvalidDataException
Thrown if the file’s header is not a valid LoRA adapter format.
- RuntimeException
Thrown if the adapter fails to preload correctly, for example due to native backend errors.
ApplyLoraAdapter(LoraAdapterSource)
Preloads a Low-Rank Adaptation (LoRA) adapter into the model using parameters from a LoraAdapterSource instance. Upon preload, the adapter is registered in the model’s Adapters collection and will only be applied at inference time when its Scale is greater than zero.
public void ApplyLoraAdapter(LoraAdapterSource loraAdapterSource)
Parameters
loraAdapterSource
LoraAdapterSourceAn instance of LoraAdapterSource containing the path to the LoRA parameters and the desired scale factor.
Remarks
Supports dynamic adapter loading: adapters are registered here and only take effect during subsequent inference if their scale is set above 0. A scale of 0 means the adapter remains inactive until the scale is raised and the model is re-evaluated.
Exceptions
- ArgumentNullException
Thrown if the
loraAdapterSource
parameter is null.- RuntimeException
Thrown if the adapter fails to preload correctly, for example due to I/O errors, invalid file format, or native backend failures when preparing the adapter for later inference.