By

Eric Schles

Hello and welcome to my book! You'll find the following sections:

1. Descriptive Statistics and Hypothesis testing
2. Applied Statistical Tests - A/B testing
3. Regression Introduction
4. Classification Introduction
5. Information Theory, Entropy and Tree Models
6. Neural Network Models
7. Introduction to Time Series Analysis

Each section covers about 4 to 5 chapters worth of material broken out into:

• Basics
• Mathematical Intuition
• Implementation
• Typical API
• Advanced Use Cases

In addition to the main chapters, I've added a number of 'engineering' focused chapters that are somewhat supplemental:

Sections to come:

• Reinforcement Learning
• Engineering for Data Science
• Text Processing
• Image Processing
• Support Vector Machines
• Genetic Algorithms
• neural network optimizers
• Recommender Systems
• A/B testing and other related workflows
• SQL best practice
• Timeseries Forecasting and Analysis
• Geospatial Analysis
• Geospatial and Timeseries forecasting
• Video Processing
• Building Data Dashboards
• Working With Search
• Building An OCR System
• Advanced Python Usage
• Active Learning
• Recurrent Neural Networks
• Convolutional Neural Networks
• Capsule Networks
• Adversarial Machine Learning
• Open World - in distribution out of distribution
• Bayesian Machine Learning
• Graph Based Neural Networks
• Monitoring
• Working with Spark
• Working with Streaming Data
• Ensembling - scikit learn ensembling strategies
• Random Forests
• Additive models:
  • Gradient boosted trees
  • splines
  • General Additive Models
  • adaboost
• explainability metrics
  • litany of examples
  • showing when and how they can fail
• Metrics
• Hyper parameter tunning
• Randomness in your models
• Counterfactual examples
• testing in machine learning applications

To Dos

• fix Decision Tree Implementation
• add SVM chapter
• add dimensionality reduction chapter
• add clustering chapter
• add RNN chapter
• add conv net chapter
• discuss attention
• create engineering productionization chapter
• hypothesis test as a ticket within engineering scrum context
• reproducibility of results