Toby Segaran's Programming Collective Intelligence, released originally in 2007, is still a valuable text, providing a good, dense introductory overview of the field of machine learning and statistics, with a fast pace and high signal to noise ratio. Its only problem is that the provided examples are sometimes a bit hard to follow, to say the least, thanks to the many typos, cryptic and/or inconsistent variable names, magic numbers, the lack of good formatting, and the usage of Python 2 of course.
My intention is somewhere around halfway between reworking chapters entirely and just dumping a big pile of code snippets here - in general, I aimed for an easy to follow walkthrough, commenting and/or refactoring everything that was not 100% obvious after reading the text and staring at the code for a few seconds, so that these cleaned-up examples can be used together with the book, allowing you to focus more on the core concepts. This is also a good excuse to refresh my Python skills, and catch up with the latest features - you might be able to pick up some useful tips along the way too.
For now, the very last chapter on genetic programming is available. Depending on my whim, chapter 5 (Optimization) and 6 (Document Filtering) are probable next candidates for a similar treatment.
Note: The original code examples from the book can be found on GitHub in various different repositories.