Agile Machine Learning with DataRobot: Automate each step of the machine learning life cycle, from understanding problems to delivering value

Cover
Copyright
Contributors
Table of Contents
Preface
Section 1: Foundations
Chapter 1: What Is DataRobot and Why Do You Need It?
	Technical requirements
	Data science processes for generating business value
		Problem understanding
		Data preparation
		Model development
		Model deployment
		Model maintenance
	Challenges associated with data science
	DataRobot architecture
		Hosting platform
		Data sources
		Core functions
		External interactions
		Users
	Navigating and using DataRobot features
	Addressing data science challenges with DataRobot
		Lack of good-quality data
		Explosion of data
		Shortage of experienced data scientists
		Immature tools and environments
		Black box models
		Bias and fairness
	Summary
Chapter 2: Machine Learning Basics
	Data preparation
		Supervised learning dataset
		Time series datasets
		Data cleansing
		Data normalization and standardization
		Outliers
		Missing values
		Category encoding
		Consolidate categories
		Target leakage
		Term-document matrix
		Data transformations
		Collinearity checks
		Data partitioning
	Data visualization
	Machine learning algorithms
		Unsupervised learning
		Reinforcement learning
		Ensemble/blended models
		Blueprints
	Performance metrics
	Understanding the results
		Lift chart
		Confusion matrix (binary and multiclass)
		ROC
		Accuracy over time
		Feature impacts
		Feature Fit
		Feature Effects
		Prediction Explanations
		Shapley values
	Summary
Chapter 3: Understanding and Defining Business Problems
	Understanding the system context
	Understanding the why and the how
		Process diagrams
		Interaction diagrams
		State diagrams
		Causal diagrams
	Getting to the root of the business problem
	Defining the ML problem
	Determining predictions, actions, and consequences for Responsible AI
	Operationalizing and generating value
	Summary
	Further reading
Section 2: Full ML Life Cycle with DataRobot: Concept to Value
Chapter 4: Preparing Data for DataRobot
	Technical requirements
		Automobile Dataset
		Appliances Energy Prediction Dataset
	Connecting to data sources
	Aggregating data for modeling
	Cleansing the dataset
	Working with different types of data
	Engineering features for modeling
	Summary
Chapter 5: Exploratory Data Analysis with DataRobot
	Data ingestion and data cataloging
	Data quality assessment
	EDA
	Setting the target feature and correlation analysis
	Feature selection
	Summary
Chapter 6: Model Building with DataRobot
	Configuring a modeling project
	Building models and the model leaderboard
	Understanding model blueprints
	Building ensemble models
	Summary
Chapter 7: Model Understanding and Explainability
	Reviewing and understanding model details
	Assessing model performance and metrics
	Generating model explanations
	Understanding model learning curves and trade-offs
	Summary
Chapter 8: Model Scoring and Deployment
	Scoring and prediction methods
	Generating prediction explanations
	Analyzing predictions and postprocessing
	Deploying DataRobot models
	Monitoring deployed models
	Summary
Section 3: Advanced Topics
Chapter 9: Understanding Forecasting and Time Series Modeling
	Technical requirements
		Appliances energy prediction dataset
	Conceptual introduction to time series forecasting modeling
	Defining and setting up time series projects
	Building time series forecasting models and understanding their model outcomes
	Making predictions with time series models
	Advanced topics in time series modeling
	Summary
Chapter 10: Recommender Systems
	Technical requirements
		Book-Crossing dataset
	A conceptual introduction to recommender systems
	Approaches to building recommender systems
		Collaborative filtering recommender systems
		Content-based recommender systems
		Hybrid recommender systems
	Defining and setting up recommender systems in DataRobot
	Building recommender systems in DataRobot
	Making recommender system predictions with DataRobot
	Summary
Chapter 11: Working with Geospatial Data, NLP, and Image Processing
	Technical requirements
		House Dataset
	A conceptual introduction to geospatial, text, and image data
		Geospatial AI
		Natural language processing
		Image processing
	Defining and setting up multimodal data in DataRobot
	Building models using multimodal datasets in DataRobot
	Making predictions using a multimodal dataset on DataRobot
	Summary
Chapter 12: DataRobot Python API
	Technical requirements
		Automobile Dataset
	Accessing the DataRobot API
	Using the DataRobot Python client
		Programming in Python using the Jupyter IDE.
	Building models programmatically
	Making predictions programmatically
	Summary
Chapter 13: Model Governance and MLOps
	Technical requirements
		Book-Crossing dataset
	Governing models
	Addressing model bias and fairness
	Implementing MLOps
	Notifications and changing models in production
	Summary
Chapter 14: Conclusion
	Finding out additional information about DataRobot
	Future of automated machine learning
	Future of DataRobot
Other Books You May Enjoy
Index