Readings

• Main Textbook (FCML) Homepage
• Bayesian Reasoning and Machine Learning (BRML; Barber) (pdf, homepage)
• Chapter 1 of Machine Learning: A Probabilistic Perspective (Murphy)

Companion Code

• FCML Example MATLAB/Octave Code and Data (Dropbox)
• FCML Example MATLAB/Octave and (limited) Python Code (GitHub)
• BRML Example Code (MATLAB/Octave)
• BRML Example Code (MATLAB/Octave)

Math Resources

Linear Algebra

• Reference for definitions of various matrix operations

Calculus and Numerical Computing

• Chapter 4 from the Deep Learning textbook

Python Resources

Getting Set Up

• Getting Started with Python for Science
  • You may want to start here, even if you’ve used Python before, as doing scientific computing can feel rather different than using Python for generic coding tasks. It walks you through getting a Scientific Computing “ecosystem” set up before going over language basics, and then reviewing the key functions in the core scientific computing libraries
• Anaconda (recommended if you are new to Python)
  • A self-contained Python distribution that helps with installation and management of libraries like numpy, scipy, pandas and matplotlib
• Spyder IDE
  • An “interactive development environment”; basically, a fancy text editor that understands Python code (analogous to what RStudio does for R if you’ve used that)
• PyCharm
  • An alternative IDE (a commercial product but with a free version). Pick whichever you prefer.

Reference and More Detailed Information

• Python Homepage
• Python Quick-Reference (with representative code samples)
• A fairly in-depth self-teaching “book” that starts from scratch
  • If you want a more thorough and detailed “course” on coding in Python, you could use this, though I’m hoping that most people in this course will not need this much detail
• Getting Started with numpy
  • Overlaps with the “Getting Started” link above, but a bit more depth into numpy specifically.