_The course syllabus and schedule may be revised during the semester. The latest version is always online here, at_ http://www.cim.mcgill.ca/~derek/ecse343.html Requirements ================================================================================ We outline some high-level requirements and expectations, below. Prerequisites ----------------------------------------------- * [*ECSE 205*: Probability and Statistics for Engineers](https://www.mcgill.ca/study/2019-2020/courses/ecse-205) and * [*COMP 250*: Introduction to Computer Science](https://www.mcgill.ca/study/2019-2020/courses/comp-250) * [*MATH 263*: Ordinary Differential Equations for Engineers](https://www.mcgill.ca/study/2019-2020/courses/math-263) We expect you to be comfortable with multivariate calculus, probability (discrete and continuous) and linear algebra: much of the theory we present in the course will focus on _re-framing_ mathematical constructs through an algorithmic and computational lens. Furthermore, you'll be expected to code your own assignment solutions during this course, and so comfort with data structures and algorithms will be valuable; coding solutions will not require large software engineered solutions, but will rather focus on implementing and testing the core algorithms and numerical methods. This course will afford you the opportunity to become comfortable with vectorized numerical computation using `Python` and `numpy`. Textbook --------------------------------------------------- When appropriate, supplemental recommended readings will be provided during lecture across the following textbooks. !!! Tip You don't __need__ to buy any of these texts to be successful in the course, but some of you may find it helpful. Leveraging *lecture* and *tutorial* time (i.e., consistent attendance, taking notes and asking questions) is much more important for building an understanding and retaining the knowledge you gain. *Scientific Computing: An Introductory Survey* (_Revised 2nd ed._)
by Michael T. Heath ![Recommended Text I](https://my.siam.org/Store/ProductImages/CL80large.jpg width="25%") This text focuses on the core ideas behind fundamental algorithms, rather than a detailed numerical analysis. Problems in mathematical & computational modeling, data analysis, and hands-on engineering examples are presented in a clear and concise style. Readers are guided through the processes of properly formulating any specific problem, carefully choosing an appropriate numerical algorithm to resolve the problem, and interpreting the results of the algorithm. Our course will rely heavily on the exposition and examples provided in _Scientific Computing: An Introductory Survey (Revised Second Edition)_. *Numerical Algorithms:* Methods for Computer Vision, Machine Learning, and Graphics
by Justin Solomon ![Recommended Text II](https://people.csail.mit.edu/jsolomon/assets/bookcover.jpg width="25%")The core concepts in numerical methods for linear and non-linear problems are presented from a similar perspective, with insightful geometric interpretations and diagrams. What most differentiates this book is a bit more of a focus on methods employed in current practice, presenting a more hands-on approach to numerical analysis for modern computer scientists and engineers. Examples vary across many computational tasks, from data processing, computational photography, and physics-based animation. The text does a good job guiding the reader from theoretical insights to numerical modeling and algorithmic design. As with any modern text on the subject, this one covers topics from numerical linear algebra to optimization and differential equations. The text targets an advanced undergraduate and graduate audience, assuming strong foundations in calculus and linear algebra. ![Recommended Text III](https://my.siam.org/Store/ProductImages/CS07large.jpg width="25%") *A First Course in Numerical Methods*
by Uri M. Ascher and Chen Greif This text also balances theoretical concepts with practical knowledge, doing a good job of highlighting the importance of moving from a mathematical formulation of a problem to a computational one. I particularly like the way this text presents the importance of parallelization, cache coherence, and other systems-level details that must be treated by a _designer_ of efficient numerical methods. That being said, those with less experience in advanced undergraduate linear algebra might have a harder time working through some of the more mathematical concepts presented early on the book; that being said, in taking the time to build this background, one becomes much better equipped to work through the rest of the "standard set" of numerical methods presented throughout the book. This text also does a very good job of helping the reader understand the many reasons behind the success and failure of numerical software. MATLAB is treated as a first class citizen in this text, and the transition from MATLAB- to Python-based numerical method development is not a large one. The target audience, as with the Solomon text, is senior undergraduate and beginning graduate level. Equipment --------------------------------------------------- IT has configured lab machines in Trottier with an ECSE 343 development stack. The [first assignment](#Assignments) provides multi-platform instructions for setting up a local development environment, for completing work on a personal machine. Time --------------------------------------------------- Attendance is *strongly encouraged*. Plan to invest *9 hours of work per week*, on average, including lecture and tutorial time. For assignments, start early and schedule appropriately. Attending tutorials with your prepared questions is a good way to skirt avoidable difficulties. Evaluation ===================================================================== Your evaluation is based on Python coding [assignments](#assignments) and a final [group project](#project).