Build on your foundation in Large Language Models (LLMs) with practical techniques for improving data quality, fine-tuning, and deploying NLP models. This course focuses on hands-on experience with LLM DataStudio and H2O LLM Studio, guiding you through dataset preparation, model customization, and evaluation. Explore advanced topics like quantization and LoRA to optimize model performance for real-world applications. Designed for learners with prior exposure to LLMs, this course helps deepen your understanding and awards you a certification upon completion.
What you'll learn
- Data Quality for NLP Models
Importance of clean data
Data preparation examples
- LLM DataStudio for Data Prep
Supported workflows
Interface exploration
Workflow overview and customisation
Quality control techniques
Setting up projects
Collaboration features
- QnA Dataset Preparation
Creating QnA datasets
Validation and quality assurance
- LLM Fine-Tuning Benefits
- Fine-Tuning with H2O LLM Studio
Introduction to H2O LLM Studio
Fine-tuning workflows
Selecting task-specific data
Data augmentation techniques
Choosing model architectures
Leveraging pre-trained models
- Quantisation and LoRA
Model compression techniques
- LLM Certification Level 2:
An advanced certification showcasing your expertise in data preparation, fine-tuning, and optimisation of language models. This certification is invaluable for professionals seeking to master the intricacies of language model workflows, enabling them to excel in specialised roles in natural language processing, machine learning, and data engineering.