Bøger udgivet af Cayenne Canyon Press

In Part III of this series, we cover the fundamentals of machine learning, focusing on:validation methodology (reprint)nearest neighbor, k-means, support vector machines, principal component analysistree-based methods: decision trees, bagging, random forest, boosting, XGBoostartificial neural networks and deep learningreinforcement learningThe focus is on algorithmic development and programming. We code each technique from scratch in Python, using an object-oriented approach.

In Part II of this series, we cover the elements of statistical modeling, focusing on:validation methodologyprinciples of object-oriented designlinear and logistic regressiongeneralized linear modelscausalitytime series analysisBayesian statistics, including simulations in pymc3Modeling customer lifetime values, including a detailed study of the beta-Bernoulli/beta-binomial model, a discretized version of the classic Pareto/NBDan introduction to credibility theoryThe theory is illustrated with simulations in Python throughout the text.

In Part I of this series, we cover basic statistical inference and experimentation, focusing on: basic statistics;derivation and review of key distributions and their relations;hypothesis testing, including an in depth power analysis for the chi-squared statistic;experimentation, including A/B tests, stratification, one- and two-factor experiments, and an introduction to bandit algorithms; maximum likelihood;gradient descent;introduction to survival analysis and stochastic processes, including empirical estimation of online survival and event processes.The theory is illustrated with simulations in Python throughout the text.