Uses of Record Class
io.github.kirstenali.deepj.training.TrainingProgress
Packages that use TrainingProgress
-
Uses of TrainingProgress in io.github.kirstenali.deepj.persistence
Methods in io.github.kirstenali.deepj.persistence that return TrainingProgressModifier and TypeMethodDescriptionstatic TrainingProgressTrainingCheckpoint.load(DecoderOnlyModel model, AdamW optimizer, StatefulBatchSource dataset, CosineLearningRateSchedule schedule, Path path) static TrainingProgressTrainingCheckpoint.load(List<Parameter> params, AdamW optimizer, StatefulBatchSource dataset, CosineLearningRateSchedule schedule, Path path) Methods in io.github.kirstenali.deepj.persistence with parameters of type TrainingProgressModifier and TypeMethodDescriptionstatic voidTrainingCheckpoint.save(DecoderOnlyModel model, AdamW optimizer, StatefulBatchSource dataset, TrainingProgress progress, CosineLearningRateSchedule schedule, Path path) static voidTrainingCheckpoint.save(List<Parameter> params, AdamW optimizer, StatefulBatchSource dataset, TrainingProgress progress, CosineLearningRateSchedule schedule, Path path) -
Uses of TrainingProgress in io.github.kirstenali.deepj.training
Methods in io.github.kirstenali.deepj.training that return TrainingProgressMethods in io.github.kirstenali.deepj.training with parameters of type TrainingProgressModifier and TypeMethodDescriptionTrainer.train(int maxSteps, int batchSize, int logEvery, float emaBeta, Float targetEmaLoss, int releaseEverySteps, Trainer.StepHook stepHook, TrainingProgress progress)