Lecturers: Erik Schnetter, Dustin Lang, several guest lecturers
Outline: This course is a practical introduction to computational physics, mixing theoretical lectures and lab-based (programming) work. The main topics will be:
- Hyperbolic partial differential equations (PDEs) with examples from general relativity
- Linear algebra (matrix factorization) with examples from condensed matter physics
- Optimization (in particular convex optimization) with examples from quantum information
- Data analysis (image processing) with examples from astronomy
The course begins with an introduction to the Julia language and related practices (e.g. version control, automated tests). It covers also certain software technologies such multi-threaded, distributed, or GPU programming.
Reference material: no textbook, although "Scientific Computing" by Heath might be useful
Course number: This is course PHYS 776 at the University of Waterloo
Format: The course consists of six two-week modules; most can be attended independently
Assessment mechanism: One lab assignment per module, pair work is encouraged
Assessment type: pass/fail
Previous Location: Time Room, Perimeter Institute
Location: Online via Zoom; details see below
Time: Mondays 13:30 - 15:00 and Wednesdays 12:30 - 14:00
First Lecture: Monday Jan. 13, 2020
Contact Erik Schnetter [email protected] by email, or open an issue in this repository (which will be public).
Week 1: Jan 13 & 16: (Note: Class meets on Thu instead of Wed this week!) Introduction to the Julia programming language. Most of the course content will be based on Julia.
Week 2: Jan 20 & 22: Reproducible science: Git, Github and shell
scripts. Version control with git and friends is a must these days,
not just for big software projects, but also for small lab projects,
coursework, and papers. Being able to reproduce your results is also
essential, and writing bash shell scripts is a reasonable way of
doing this.
-
In preparation for the Jan 20 lecture, please obtain a Github account (instructions here.)
Week 3: Jan 27 & 29: No lectures (PSI Winter school)
Week 4: Feb 3 & 5: Discretizing functions, (elliptic) partial differential equations (PDEs). One often needs to represent functions in a computer (e.g. a density or velocity field). We discuss and experiment with a few approaches. We also discuss elliptic PDEs, and will then calculate the structure of a spherically symmetric star in general relativity, solving the Tolman-Oppenheimer-Volkoff (TOV) equation.
Week 5: Feb 10 & 12: Time-dependent (hyperbolic) partial differential equations (PDEs). Time-dependent problems are more difficult to solve, as there are numerous ways to encounter instabilities. We will discuss topics such as well-posedness, stable discretizations, and will solve a wave equation.
Week 6: Feb 17: No lecture (Family Day)
Week 6: Feb 19: (Reading week) General Q&A session, no fixed topic
Week 7: Feb 24 & 26: Spectral representations. Working in Fourier space or with Spherical Harmonics is more natural for some problems, and the Fast Fourier Transform (FFT) makes this computationally tractable. FFT libraries have some complexities (pun intended), so we will experiment with them.
Week 8: Mar 2 & 4: Will East: Relativistic Hydrodynamics
Week 9: Mar 9 & 11: Denis Rosset: Convex Optimization
Week 10: Mar 16 & 18: Parallel Computing (threads), Distributed Computing (MPI)
- **This lecture and the following will be taught online via Zoom: Monday: https://pitp.zoom.us/j/182113370, Wednesday: https://pitp.zoom.us/j/956012168 **
Week 11: Mar 23 & 25: Image Processing I. Using astronomical images as an example, we will experiment with some basic image processing algorithms.
Week 12: Mar 30 & Apr 1: Image Processing II. Building on last week's experiments, we will work on detecting and measuring stars in astronomical images.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent