AstroML Interactive Book

astroML is a Python module for machine learning and data mining that accompanies the book “Statistics, Data Mining, and Machine Learning in Astronomy”, by Željko Ivezić, Andrew Connolly, Jacob Vanderplas, and Alex Gray. astroML is built on numpy, scipy, scikit-learn, matplotlib, and astropy, and contains a growing library of statistical and machine learning routines for analyzing astronomical data.

In this interactive book we provide notebooks that describe the statistical and machine learning methods used in astroML together with code that runs these methods on existing astronomical data sets. The structure of this interactive book follows the chapters of “Statistics, Data Mining, and Machine Learning in Astronomy”. Each notebook can viewed through the browser (with navigation links at the side of the page), be downloaded to your own computer, or be executed directly using Binder or Google Colab

Content

• Chapter 1: Introduction and Data Sets
• Chapter 2: Fast Computation and Massive Datasets
• Chapter 3: Probability and Statistical Distributions
  • Overview of Probability and Random Variables
  • Descriptive Statistics
  • Univariate Distribution Functions
  • The Central Limit Theorem
  • Bivariate and multivariate distribution functions
• Chapter 4: Classical Statistical Inference
  • Maximum Likelihood Estimation (MLE)
  • MLE applied to Gaussian mixtures
• Chapter 5: Bayesian Statistical Inference
  • Parameter Estimation for a Gaussian Distribution
  • Parameter Estimation for a Binomial Distribution
  • Parameter estimation for the Cauchy (Lorentzian) distribution
  • Approximate Bayesian Computation Example
  • Hierarchical Bayes Example
• Chapter 6: Searching for Structure in Point Data
  • Density Estimation for SDSS “Great Wall”
  • Searching for Structure in Point Data
  • Gaussian Mixture Models Example
  • Extreme Deconvolution
• Chapter 7: Dimensionality and its Reduction
  • Dimensionality reduction
• Chapter 8: Regression and Model Fitting
  • Measurement Errors in Linear Regression
  • Measurement errors in both dependent and independent variables
• Chapter 9: Classification
  • Classification
  • Deep Learning: Classifying Astronomical Images
• Chapter 10: Time Series Analysis
  • Modeling Toolkit For Time Series Analysis
  • Wavelets
  • Digital Filtering
  • Temporally localized signals
  • Matched Filter Chirp Search
  • Analysis of Stochastic Processes

Contributing