# Assignment 5

Assignment 5 is now available! There is no standalone document for this assignment, just a code skeleton in the class repository and section 8.1 of the class notes. You do not need to submit the exercises in the text, only the programming component. This assignment will be due at 11:00 pm on March 28th. The submission guidelines are identical to the previous assignments. Electronic submissions are preferred (cc’ing both kmcrane@cs.cmu.edu and nsharp@cs.cmu.edu).

Additionally, include in your email a paragraph or two commenting on the results of this assignment. If your implementation is successful, analyze the results and comment on why they make sense. If you do not have faith in your results, comment on what appears to be wrong and what may be the cause.

## 3 thoughts on “Assignment 5”

1. Slav says:

In section 8.1 of the notes, it mentions that system solvers ‘solvePositiveDefinite’ and ‘solveSquare” are provided that we can use. Are these indeed provided? If so, how should they be called? (they were not found when I tried to call them)

1. Khryl says:

@Slav, I’m not using these two methods. I just use scipy.sparse.linalg.spsolve to solve the problem. It’s usually pretty fast.

1. ^^^ This is the right idea! Those method names referenced in the text correspond to a previous version of this codebase, we’ll just be using the methods in scipy.sparse.

If you find yourself wanting anything else you can find a thorough list of the available routines here http://docs.scipy.org/doc/scipy/reference/sparse.linalg.html .

To call them, you’ll want to use something like scipy.sparse.linalg.spsolve(arguments).