Fine-Tuning LLMs: LoRA, QLoRA, and Full Fine-Tuning
«Fine-Tuning LLMs: LoRA, QLoRA, and Full Fine-Tuning» è un corso online di RAG e AI engineering di livello avanzato, tenuto da InsightLearn Curated AI su InsightLearn. Il percorso comprende 9 lezioni organizzate in 3 sezioni, per una durata complessiva di 14h 0m. Fine-tune open-source LLMs with LoRA and QLoRA: dataset preparation, training, evaluation, and deployment.
- Docente: InsightLearn Curated AI
- Categoria: RAG e AI engineering
- Livello: avanzato
- Durata: 14h 0m
- Accesso: gratuito in questa fase di validazione, in cambio di un feedback al termine
- Certificato: certificato di completamento verificabile online
Descrizione del corso
InsightLearn è in fase di sviluppo (work in progress): i contenuti di questa pagina sono pubblicati esclusivamente a fini di verifica e test.
A practical course on parameter-efficient fine-tuning. Covers when to fine-tune vs prompt engineer, dataset curation for instruction tuning, LoRA fundamentals (rank, alpha, target modules), QLoRA with 4-bit quantization, supervised fine-tuning with HuggingFace TRL, DPO and RLHF alignment, LoRA adapter merging, evaluation on domain benchmarks, and deploying fine-tuned models with Ollama and vLLM.
Cosa imparerai
- Curated from public resources. Fine-tune open-source LLMs with LoRA and QLoRA: dataset preparation, training, evaluation, and deployment.
Programma del corso
- When to fine-tune vs prompt engineer (25 min)
- Dataset curation for instruction tuning (35 min)
- LoRA mathematics: low-rank decomposition explained (30 min)
- QLoRA: 4-bit quantization + LoRA on consumer GPUs (45 min)
- SFTTrainer: supervised fine-tuning with TRL (40 min)
- DPO: Direct Preference Optimization (38 min)
- Merging LoRA adapters and pushing to HuggingFace Hub (25 min)
- Evaluating fine-tuned models with EleutherAI harness (30 min)
- Deploying fine-tuned models with Ollama and vLLM (28 min)
Dalle lezioni
Estratto della trascrizione di «When to fine-tune vs prompt engineer»
Welcome to this lecture of the end-to-end course of LRMS. Lo-ka, ku-lo-ka, and full-fine-to-ing. In this session, we will explore the end-to-end WS Prompt Engineer. We will start with the fundamental concepts and the basic theoretical principles. We will then see practical applications and practical cases in the world of development. We will analyze concrete examples of practical cases. At the end of this lesson you will be able to apply autonomously the techniques of WENTO until you complete your projects. We will start with the fundamental concepts and the basic theoretical principles. We will then see practical applications and almost real cases in the world of AI development. We will analyze concrete examples and the best practices in the…
Estratto della trascrizione di «Dataset curation for instruction tuning»
Benvenuto in questa lezione del corso fine tuning LLMS. Lo rà, cu lo rà, and full fine tuning. In questa session esploveremo data set curation for instruction tuning. Inizieremo con i concetti fondamentali e i principi teorici di base. Vedremo poi applicazioni pratiche e casi d'uso reali nel mondo dello sviluppo AI. Analizzeremo e i concetti elementi pratiche d'uso reali. At the end of this lesson, you will be able to apply the data set curation for instruction tuning in your projects. We will start with the fundamental concepts and the basic theoretical principles. We will then see practical and almost universal applications in the world of development. We will analyze concrete examples and the best practices in the sector.…
Estratto della trascrizione di «LoRA mathematics: low-rank decomposition explained»
Welcome to this lecture of the last course of tuning of LLMS, LOKA, QLOKA, and full tuning. In this session, we will explore the mathematics, the composition, and explain it. We will start with the fundamental concepts and the theoretical principles of the base. We will then see practical applications and almost universal applications in the world of development. We will analyze concrete elements of mine. and the practical of the sector. At the end of this lesson, you will be able to apply autonomously the techniques of the mathematics program in your projects. We will start with the fundamental concepts and the theoretical principles of the base. We will then see practical applications and almost real-life cases in the…
Altri corsi di RAG e AI engineering
- Function Calling & Tool Use: Multi-Model Production Patterns
- LangSmith: Tracing, Debugging, and Monitoring LLM Apps
- LLM Evals: RAGAS, DeepEval, and Automated Testing
- Local LLMs: Ollama, vLLM, and Self-Hosted AI
- Multimodal AI: Vision, Audio, and Image Generation APIs
- Advanced Prompt Engineering: Chain-of-Thought, Few-Shot, and Beyond
Domande frequenti
Quanto dura il corso Fine-Tuning LLMs: LoRA, QLoRA, and Full Fine-Tuning?
Il corso dura 14h 0m in totale, con 9 lezioni suddivise in 3 sezioni, da seguire online al proprio ritmo.
Serve esperienza pregressa?
Il corso è di livello avanzato e non indica requisiti specifici di ingresso.
Il corso rilascia un certificato?
Sì: al completamento del corso ricevi un certificato di completamento verificabile online.