It’s finally here: My last semester! Due to reasons I don’t want to get into, I took on a somewhat ridiculous workload the last few semesters, and specially this semester. Here we go:
This was actually a very interesting course. It is not often taught by a sessional, but it was this time. A PhD candidate named Hassan Khosravi was teaching it, and he was by far the best sessional I’ve ever had, and better than many profs as well.
The assignments were not particularly difficult for a 3rd year class (despite what some whiners on ratemyprofessor may have to say), nor was the material (but having taken 405 already and taking 413 at the same time certainly left lots of crossover for me. I guess it would be more challenging if someone was taking this before 307 and any stats courses. The department really should work on fixing prerequisites). My only regret? We didn’t get to cover neural networks. The savior? You can get extra credit in this course by doing a presentation. Whereas most people opted to just sum up a chapter we had already seen, some of us presented new material. My friend Dave presented something on Watson (the Jeopardy robot), I did a presentation on neural networks (which forced me to research and learn about them), and my friend Rhiannon did a presentation on a cool lyric-generating project.
Either way, this was interesting and fun. I would recommend this course.
Wow. Let’s put it this way: I won’t name this professor. Those of you who have read my reviews before know that this is not a good sign. The course and the professor were both horrible. To complete the trifecta, the TA was….Well…see the blog post just before this one!
Course material: Almost useless. What would have been useful would have been some socket programming. None was done. I hear Dr.Peters (see my review of CMPT 307 a few years back) who was teaching it in Surrey at the same time did have some socket programming. Crap.
Some of the actions of this professor, along with the abundance of copied and useless assignments and projects, combined with the excrement-storm I generated because of it will most likely lead to some sort of departmental changes with regards to curriculum consistency and proper scheduling. At least, that’s what I’ve been told. The next generation will have to fight for that.
Regardless to say, do not take this course with this prof (determining who the prof is is left at a ridiculously easy exercise for the reader.)
Another very interesting, but definitely not easy course taught by Anoop (I took compilers with him last semester). This course had some really cool stuff about Natural Language Processing, and also had the added bonus (for me, at least) of teaching me Python (I totally understand the love affair with Python now). Unfortunately, Anoop is smart. Really smart. Scary smart. And he doesn’t always realize that some of us are nowhere near that smart. His assignments are often mind-breaking. There is usually a really easy way to do most parts of it, but figuring out what that may be requires either reading odd source code, being as smart as him, or bashing your head against the problem for several hours/days. Combine that with a semester where I was taking 4 upper-division courses, and that made for…a little stress.
However, the class was interesting, I like the way Anoop teaches, and, despite the sensation of having to walk funny after the assignments, most of them were at least somewhat fun. I would recommend this course.
Take this course when Greg Baker is teaching it. I’ve already sung his praises in a past review, and if I do it again, my missus is going to get jealous. Greg took the approach he developed while getting ready for 383 (see my review of that) to his new-and-improved 470. This means there were some easy labs that eased people unfamiliar with the material into it (going from ridiculously easy to a bit more involved), a technology evaluation, a poster presentation (the only miserable part of this course) and a semester-long project that is really worth the largest chunk of the grade.
I really owe my grade in this one to one of my group members, Dave, who ninja-code incessantly. Every time the other member and myself turned around, Dave had added more cool stuff. That didn’t stop us from doing a ton of work and still pulling several really long days just before the project was due to get the final bits done (mainly, the front end). I think our project was educational, and might even have some possible marketability given some more work. It allowed me to learn Django (and how they do MVC, as opposed to Rails, which I got to work on in the past) as well. I know why Greg loves Django (not to be pronounced duh-jango), but I still maintain that their documentation sucks.
Take this course with Greg, obviously.