Class FinetuningProgressEventArgs
- Namespace
- LMKit.Finetuning
- Assembly
- LM-Kit.NET.dll
Provides data for the fine-tuning progress event, including metrics and state management.
public sealed class FinetuningProgressEventArgs : EventArgs- Inheritance
-
FinetuningProgressEventArgs
- Inherited Members
Fields
- BestLoss
Gets the best loss value of the fine-tuning process, representing the training error.
- Elapsed
Gets the total elapsed time of the fine-tuning process.
- Epochs
Gets the number of completed epochs in the fine-tuning process.
- IterationCount
Gets the total number of iterations to be executed in the fine-tuning process.
- Iterations
Gets the number of completed iterations in the fine-tuning process.
- Loss
Gets the current loss value of the fine-tuning process, representing the training error.
- NextSample
Gets the index of the next sample to be processed.
- Percentage
Gets the completion percentage of the fine-tuning process.
- Remaining
Gets the estimated remaining time for the fine-tuning process.
- SampleCount
Gets the total number of samples to be processed in one epoch.
- Stop
Gets or sets a value indicating whether the fine-tuning process should stop.
- TrainedSamples
Gets the number of trained samples in the fine-tuning process.
Methods
- SaveLora(string)
Saves the current state of the LoRA adapter to the specified destination path.
- SaveLoraCheckpoint(string)
Saves a checkpoint of the current LoRA training state to the specified path.