Property EnableQuantization
- Namespace
- LMKit.Finetuning
- Assembly
- LM-Kit.NET.dll
EnableQuantization
Gets or sets a flag indicating whether quantization should be applied to the merged model's weights.
public bool EnableQuantization { get; set; }Property Value
Remarks
When true, after successfully merging LoRA adapters into the base model,
the property triggers a quantization pass to reduce model size and improve inference speed.
Quantization is only performed if the input and output precisions differ; in that case,
the merged weights are converted to F16 precision as the final output.
The default value is false, meaning quantization is disabled unless explicitly enabled.