Introduction
Welcome to this course
()
What you should know
()
1. Data Preparation for Predictive AI
Data exploration and initial quality assessment
()
Detecting and managing missing data
()
Detecting and managing outliers
()
Challenge: Assess data quality of a dataset
()
Solution: Assess data quality of a dataset
()
Feature engineering: Scaling and normalizing data
()
Feature engineering: Categorical encodings
()
Challenge: Apply feature engineering to a dataset
()
Solution: Apply feature engineering to a dataset
()
2. Data Preparation for Generative AI
Structured vs. unstructured data
()
Vector representations of unstructured data
()
Tools for generating vector representations
()
Similarity between vector representations
()
Challenge: Choosing vector generation tool
()
Solution: Choosing vector generation tool
()
3. Data Quality Assessment
Elements of data quality: Consistency, accuracy, and completeness
()
Statistical techniques for data quality assessment
()
Challenge: Data quality
()
Solution: Data quality
()
4. Data Augmentation for Generative AI
Overview of data augmentation
()
Text perturbation and style transfer
()
Retrieval-augmented generation (RAG)
()
Introduction to LangChain for RAG
()
Challenge: Understanding components of RAG
()
Solution: Understanding components of RAG
()
5. Knowledge Augmentation for Generative AI
Overview of knowledge augmentation
()
Fine-tuning and knowledge distillation tools and techniques
()
Ontologies, taxonomies, and knowledge graph tools and techniques
()
Challenge: Knowledge augmentation
()
Solution: Knowledge augmentation
()
6. Development Practices for AI Applications
Determining what data sources to use
()
Applying data quality checks
()
Building models
()
Model assessment
()
Challenge: Development practices
()
Solution: Development practices
()