Namespace LMKit.Finetuning
Namespaces
Classes
- ChatTrainingSample
Represents a single supervised fine-tuning sample for chat-based models. Encapsulates the full ChatHistory to learn from and the target InferenceModality (e.g., text-only or multimodal).
- LoraAdapter
Represents a LoRA adapter that applies low-rank adaptation weights to a language model.
- LoraAdapterSource
Represents a source for a LoRA adapter.
- LoraMerger
Handles the merging of Low-Rank Adaptation (LoRA) transformations into a base model.
This class is responsible for loading LoRA transformation files and integrating them with a base model's weights.
- TrainingDataset
Represents a supervised fine-tuning dataset composed of multiple ChatTrainingSample instances. Provides collection management and export functionality to the ShareGPT JSON format.
- TrainingSample
Represents a training sample consisting of a sequence of tokens and its decoded value.