Model training

Tutorials for training, fine-tuning, and hyperparameter optimization of models at scale.

Hyperparameter optimization

Run large-scale HPO experiments with zero manual tracking, deterministic results, and automatic recovery.

LLM fine-tuning with LoRA and QLoRA

Fine-tune a language model for SQL generation using full, LoRA, or QLoRA methods in one Flyte pipeline.

BERT emotion classification

Fine-tune ModernBERT on Twitter emotion labels with confusion-matrix evaluation and attention visualizations.