classes
Um yeah, let's see how this goes now, two months later. Classes are over now, so I finally have some time to write some stuff, though I really don't feel like doing it because this is going to be an annoying task for sure, I'm splitting up entries in rough groups, so they won't be too long.
Classes:
-6.839
So this class got a little bit ridiculous because only four of us ever showed up at any given time, mostly me, Jordan, Alan, and Tiffany. It was awkward, towards the end of the course we started discussing articles rather than having lectures, and there was one time I came in 30 minutes late, and hot professor was like, "So who wants to cancel class, given there are so few people here?" I wanted to leave, and I think we all kind of did, but at the same time I was like, ok I just got here, might as well stay... Finally Alan spoke up and we ended up sticking around and talking about the Teddy software. But all of those paper discussion classes were just us watching the videos and conference recordings off the people's websites, hot professor was getting kind of lazy I think. Which is totally okay because it was fun to watch that stuff anyway.
Since I last wrote, we had two more regular projects and one final project. The very first project that we had, the ray tracer, was fairly straightforward, so I thought everything would proceed like 837. No such luck. The next projects got hard, real fast.
-I basically spent all of spring break working on the second project, since I had put off dealing with it. The biggest challenge was reading the notes and figuring out just what the hell was going on with the math in this class. The second project had to do with physically based modeling, and we had to implement constraint forces and rigid bodies. I spent so long figuring stuff out that I never got the constraints to work after a NaN error, but I did get rigid bodies going in 3D. Turns out we only had to do them in 2D, so I got a little extra credit for that, even though I got credit off for not getting constraints working. I was even lucky at that, because it wasn't until morning of that I actually got the rigid bodies to stop spazzing out (stupid quaternion conversions). So I had gone into class and volunteered to go first because I was depressed and sure my presentation would be the worst. Luckily hot professor wasn't there, it was just Eugene, so that made me feel better. But really I felt so much better after the class ended because basically nobody else finished the project. People got various things going, from constraints to rigid bodies in 2D, but nobody got everything. Since it was customary to finish everything in 837, I got really stressed, but after seeing the lower standards in 839, I was very relieved that I wasn't much further behind everyone else.
It was also cool because we all now had this shared bonding experience because I think we were all in the same stressed situation. We were all chatting about all of our frustrations with the project, and Alan was like, "At least I finally understand Jacobians," and I couldn't have agreed more. Even though it was all very annoying, I admittedly did learn a lot of stuff, especially about math...
-Third project wasn't so bad, we were doing inverse kinematics. I never got mine working in 3D, but I was now under the more relaxed, do-as-much-as-you-can attitude, so it was much less stressful. Indeed, not so many people got everything working, though some did. At least the 2D worked, and that was pretty cool.
-Final project was severely stalled because I had dance troupe that same week and was therefore quite busy. I decided to do subdivision, because I didn't feel like going back and dealing with the previous projects and all their shortcomings. Subdivision was a pretty easy concept but it was a lot of bookkeeping and attempts at optimization. I was pretty happy with my subdivision program, but I just regret that I didn't have time to get it to add/delete new points. I might still tackle that later. I also would like to have had texture mapping of some sort. Also, as pointed out by the professor, my shading might have been off.
I was supposed to give my presentation last on Tuesday, but because a) hot professor spent so long on the first presentation asking Jay to change things in his code on the spot, and b) kenny showed up late and gave his presentation but tried to render on the spot, which was really dumb, I had to be postponed until Thursday. While that would have been nice to have extra time to finish my work, I was really counting on just being done on Tuesday and focusing my energy on my 6.865 project, which was also due Thursday. Having the two extra days gave me pressure, like I had to improve my project somehow, and that was annoying. I ended up just going with the disclaimer that I didn't have time to do anything else. I think I still had some cool features, so I tried to play those up instead.
-Final exam. It was on Monday morning at 9am, so I spent Sunday evening studying, putting together notes on all of my projects, since we were allowed to bring anything handwritten. It was lucky that I always write out detailed plans and mathematical derivations every time I did a project, so I was ahead of the game on that. The exam itself was three questions, one on each project asking for the mathematical derivation of some aspect of the project. The first one on ray tracing was easy, since I knew exactly how to do the derivation for sampling the spherical light (I even wrote notes because I suspected that would be asked). Second question dealt with a rigid body with a nail constraint, and the third was the inverse kinematics on a rigid body. I never went so detailed into rigid bodies other than just moving them, so I had no idea what to do. I ended up just doing my best with a guess at a possible approach. I'm not sure if that was right or wrong, but I got a 91 on the final, so I guess it was not bad.
I was really happy when I was done with the final, I was holding office hours later that day (though only one person came), so I went to the Galleria and bought a book on Costa Rica (I'm going btw, more on that later). While I was there I ran into Kenny and Sharat, who I guess also wanted to unwind after the final. I wonder how that kid Sharat did in the class, because he even missed the final presentation because he was late and the professor needed to catch a plane. He was always really bad at all his projects, so I'm curious. But I'm happy with my grade. I think it paid off to go to all the classes and maintain a presence and show that I was interested. If we were grading on a flat 90-A scale, I would have been borderline, but I think that likely the cutoff was lower based on everyone else's performance. Anyway, turns out I can rock grad school better than undergrad, woohoo!
-6.865
I started deteriorating in this class. I still loved going to it, but I would always get sleepy halfway through (it was during normal siesta hours!) and zone out (you know when it actual becomes painful to try to force your eyes open?). I did enjoy some of the guest lecturers (Shai Avidan with seam carving, this other guy with the amplification of tiny movements, etc.), and I really liked being able to follow along and ask questions. I wish I was like that in all classes, and it really makes me realize that I like this stuff better than biology, because I never wanted to ask questions in any of my bio classes. The projects were great, and I got perfect scores on all of them, and I got to do cool things like the Jaime-Ling face warp sequence and panorama stitching... Then came the final project.
I did so badly on that final project. I'm so mad at myself, because I really liked making a good impression on Fredo. I chose a project that was hard to begin with, exacerbated by the fact that I really do not understand math and don't know anything about deconvolution. The thing was, I loved doing the projects in class because we took a concept from a paper and implemented it. Then he asked people in class for their topics of choice, and it seemed like everyone wanted to do some sort of fancy take-a-paper-and-extend-it sort of thing, where I didn't feel like I was at a point yet to be creative with that. So I thought I had to be doing actual research of some type, so I decided to look at removing blur caused by camera shake. Turns out that's really hard and the algorithm that I was looking at even Fredo didn't understand fully. Unfortunately, he didn't get back to me on my proposal until two weeks later, so I didn't know this opinion until late in the game. I wanted to do something else a little bit more fun, but I somehow got trapped in doing this, and I did the thing that I always do by waiting until it's too late to ask or turn back because I'm too embarrassed that I've waited so long. To top it off, I had no time to work on it because I knew I had to do well in my 839 project, since I was doing worse in that class. I was lucky to have gotten the code used in the paper to play around with, which allowed me some basis for working, otherwise I would have truly been in despair. I utterly failed, but was lucky enough to have gotten a 7/10 from Fredo and therefore maintained an A- in the class (too bad, could have been an A otherwise). Looked from the 2 minute class presentations that the range of success from people in the class went from around my level to really cool success, so I didn't feel that bad about it afterwards. I think I would like to try some of the other things I was interested in though, if I have time sometime.
-6.042
Glad the class is over, don't have to do any more stupid TA stuff. I don't mind it so much, but it's nice not having to get up three times a week by 9:30 and having to deal with grading and making up miniquizzes. I liked my students and all (brought candy twice too, that was fun), so I'll miss them a little. Oh, interesting thing, one of my students apparently went to the same elementary school as me, Washington school. Apparently he was good childhood friends with Eric Nellis and said that that was one of his biggest childhood inspirations, because Eric Nellis was smart, athletic, and good at video games, and that's why he chose his English name when he changed his Chinese name to Eric. Weird, small world.
Other stuff that I had to do for class, I did the magic trick again, this time as the assistant. It's much more nerve racking being the assistant though, because that's where you've gotta be really slick arranging the cards. It's easy to read them as the magician. I do enjoy doing it though, it's a good time. I also gave a mini-review lecture during the last week when the professor wasn't there. I freaked out last minute when I didn't have the slides, only the pdf, so I called the prof late at night while he was in Cali, which I felt bad about, because turns out the slides were in the restricted directory anyway. The mini lecture went fine, though it was very short and casual. Thank goodness I didn't have to do full lectures like Jay and Jeremy.
Final exam prep was again done primarily by Jay, and I felt bad about that, but that guy is seriously just such a good TA, and I didn't really have any time so god bless him. Then we gave the final exam, and I went to LaVerde's to get my first ever breakfast sandwich, and I saw one of my students in there scribbling stuff on his cheat sheet. I thought that was kind of dumb, since you'd probably get more out of the hour in the exam that you just wasted, but I guess that was his choice.
Grading the exams took again around 9 hours, which is ridiculous. Last term I only graded like one question because there were a ton of exams and that one question took so long that I got totally burned out. This time I flew through the questions, I graded like five of them and just zipped right through, that was fun, made me feel productive. We also ordered this really spicy chinese food from some place that I guess was Szechuan, and Prof Meyer told us some story about how when he first dated his wife he was impressed by how well she kept up with spicy food. He also said she used to hate him because he was on her thesis oral panel or whatever and gave her a lot of shit because he didn't like the project that her supervisor gave. It's weird to think of him ever being young, he's just so old... That food was really spicy. Really really spicy...
Grades meeting was only 3 hours this time, thank goodness. Jay had a cool projected trends in miniquizzes and hw over the semester for every student, so that was helpful. The boundaries were also pretty self defining this time, which helped. I was happy almost all of my students did pretty well, so I got a good bunch this time.
That's that for that class. I'll probably have to TA again next semester, but hopefully Jay will be there still, forever one-up-ing me in TA semesters so that I never have to take on the head TA role, ever. =P
Classes:
-6.839
So this class got a little bit ridiculous because only four of us ever showed up at any given time, mostly me, Jordan, Alan, and Tiffany. It was awkward, towards the end of the course we started discussing articles rather than having lectures, and there was one time I came in 30 minutes late, and hot professor was like, "So who wants to cancel class, given there are so few people here?" I wanted to leave, and I think we all kind of did, but at the same time I was like, ok I just got here, might as well stay... Finally Alan spoke up and we ended up sticking around and talking about the Teddy software. But all of those paper discussion classes were just us watching the videos and conference recordings off the people's websites, hot professor was getting kind of lazy I think. Which is totally okay because it was fun to watch that stuff anyway.
Since I last wrote, we had two more regular projects and one final project. The very first project that we had, the ray tracer, was fairly straightforward, so I thought everything would proceed like 837. No such luck. The next projects got hard, real fast.
-I basically spent all of spring break working on the second project, since I had put off dealing with it. The biggest challenge was reading the notes and figuring out just what the hell was going on with the math in this class. The second project had to do with physically based modeling, and we had to implement constraint forces and rigid bodies. I spent so long figuring stuff out that I never got the constraints to work after a NaN error, but I did get rigid bodies going in 3D. Turns out we only had to do them in 2D, so I got a little extra credit for that, even though I got credit off for not getting constraints working. I was even lucky at that, because it wasn't until morning of that I actually got the rigid bodies to stop spazzing out (stupid quaternion conversions). So I had gone into class and volunteered to go first because I was depressed and sure my presentation would be the worst. Luckily hot professor wasn't there, it was just Eugene, so that made me feel better. But really I felt so much better after the class ended because basically nobody else finished the project. People got various things going, from constraints to rigid bodies in 2D, but nobody got everything. Since it was customary to finish everything in 837, I got really stressed, but after seeing the lower standards in 839, I was very relieved that I wasn't much further behind everyone else.
It was also cool because we all now had this shared bonding experience because I think we were all in the same stressed situation. We were all chatting about all of our frustrations with the project, and Alan was like, "At least I finally understand Jacobians," and I couldn't have agreed more. Even though it was all very annoying, I admittedly did learn a lot of stuff, especially about math...
-Third project wasn't so bad, we were doing inverse kinematics. I never got mine working in 3D, but I was now under the more relaxed, do-as-much-as-you-can attitude, so it was much less stressful. Indeed, not so many people got everything working, though some did. At least the 2D worked, and that was pretty cool.
-Final project was severely stalled because I had dance troupe that same week and was therefore quite busy. I decided to do subdivision, because I didn't feel like going back and dealing with the previous projects and all their shortcomings. Subdivision was a pretty easy concept but it was a lot of bookkeeping and attempts at optimization. I was pretty happy with my subdivision program, but I just regret that I didn't have time to get it to add/delete new points. I might still tackle that later. I also would like to have had texture mapping of some sort. Also, as pointed out by the professor, my shading might have been off.
I was supposed to give my presentation last on Tuesday, but because a) hot professor spent so long on the first presentation asking Jay to change things in his code on the spot, and b) kenny showed up late and gave his presentation but tried to render on the spot, which was really dumb, I had to be postponed until Thursday. While that would have been nice to have extra time to finish my work, I was really counting on just being done on Tuesday and focusing my energy on my 6.865 project, which was also due Thursday. Having the two extra days gave me pressure, like I had to improve my project somehow, and that was annoying. I ended up just going with the disclaimer that I didn't have time to do anything else. I think I still had some cool features, so I tried to play those up instead.
-Final exam. It was on Monday morning at 9am, so I spent Sunday evening studying, putting together notes on all of my projects, since we were allowed to bring anything handwritten. It was lucky that I always write out detailed plans and mathematical derivations every time I did a project, so I was ahead of the game on that. The exam itself was three questions, one on each project asking for the mathematical derivation of some aspect of the project. The first one on ray tracing was easy, since I knew exactly how to do the derivation for sampling the spherical light (I even wrote notes because I suspected that would be asked). Second question dealt with a rigid body with a nail constraint, and the third was the inverse kinematics on a rigid body. I never went so detailed into rigid bodies other than just moving them, so I had no idea what to do. I ended up just doing my best with a guess at a possible approach. I'm not sure if that was right or wrong, but I got a 91 on the final, so I guess it was not bad.
I was really happy when I was done with the final, I was holding office hours later that day (though only one person came), so I went to the Galleria and bought a book on Costa Rica (I'm going btw, more on that later). While I was there I ran into Kenny and Sharat, who I guess also wanted to unwind after the final. I wonder how that kid Sharat did in the class, because he even missed the final presentation because he was late and the professor needed to catch a plane. He was always really bad at all his projects, so I'm curious. But I'm happy with my grade. I think it paid off to go to all the classes and maintain a presence and show that I was interested. If we were grading on a flat 90-A scale, I would have been borderline, but I think that likely the cutoff was lower based on everyone else's performance. Anyway, turns out I can rock grad school better than undergrad, woohoo!
-6.865
I started deteriorating in this class. I still loved going to it, but I would always get sleepy halfway through (it was during normal siesta hours!) and zone out (you know when it actual becomes painful to try to force your eyes open?). I did enjoy some of the guest lecturers (Shai Avidan with seam carving, this other guy with the amplification of tiny movements, etc.), and I really liked being able to follow along and ask questions. I wish I was like that in all classes, and it really makes me realize that I like this stuff better than biology, because I never wanted to ask questions in any of my bio classes. The projects were great, and I got perfect scores on all of them, and I got to do cool things like the Jaime-Ling face warp sequence and panorama stitching... Then came the final project.
I did so badly on that final project. I'm so mad at myself, because I really liked making a good impression on Fredo. I chose a project that was hard to begin with, exacerbated by the fact that I really do not understand math and don't know anything about deconvolution. The thing was, I loved doing the projects in class because we took a concept from a paper and implemented it. Then he asked people in class for their topics of choice, and it seemed like everyone wanted to do some sort of fancy take-a-paper-and-extend-it sort of thing, where I didn't feel like I was at a point yet to be creative with that. So I thought I had to be doing actual research of some type, so I decided to look at removing blur caused by camera shake. Turns out that's really hard and the algorithm that I was looking at even Fredo didn't understand fully. Unfortunately, he didn't get back to me on my proposal until two weeks later, so I didn't know this opinion until late in the game. I wanted to do something else a little bit more fun, but I somehow got trapped in doing this, and I did the thing that I always do by waiting until it's too late to ask or turn back because I'm too embarrassed that I've waited so long. To top it off, I had no time to work on it because I knew I had to do well in my 839 project, since I was doing worse in that class. I was lucky to have gotten the code used in the paper to play around with, which allowed me some basis for working, otherwise I would have truly been in despair. I utterly failed, but was lucky enough to have gotten a 7/10 from Fredo and therefore maintained an A- in the class (too bad, could have been an A otherwise). Looked from the 2 minute class presentations that the range of success from people in the class went from around my level to really cool success, so I didn't feel that bad about it afterwards. I think I would like to try some of the other things I was interested in though, if I have time sometime.
-6.042
Glad the class is over, don't have to do any more stupid TA stuff. I don't mind it so much, but it's nice not having to get up three times a week by 9:30 and having to deal with grading and making up miniquizzes. I liked my students and all (brought candy twice too, that was fun), so I'll miss them a little. Oh, interesting thing, one of my students apparently went to the same elementary school as me, Washington school. Apparently he was good childhood friends with Eric Nellis and said that that was one of his biggest childhood inspirations, because Eric Nellis was smart, athletic, and good at video games, and that's why he chose his English name when he changed his Chinese name to Eric. Weird, small world.
Other stuff that I had to do for class, I did the magic trick again, this time as the assistant. It's much more nerve racking being the assistant though, because that's where you've gotta be really slick arranging the cards. It's easy to read them as the magician. I do enjoy doing it though, it's a good time. I also gave a mini-review lecture during the last week when the professor wasn't there. I freaked out last minute when I didn't have the slides, only the pdf, so I called the prof late at night while he was in Cali, which I felt bad about, because turns out the slides were in the restricted directory anyway. The mini lecture went fine, though it was very short and casual. Thank goodness I didn't have to do full lectures like Jay and Jeremy.
Final exam prep was again done primarily by Jay, and I felt bad about that, but that guy is seriously just such a good TA, and I didn't really have any time so god bless him. Then we gave the final exam, and I went to LaVerde's to get my first ever breakfast sandwich, and I saw one of my students in there scribbling stuff on his cheat sheet. I thought that was kind of dumb, since you'd probably get more out of the hour in the exam that you just wasted, but I guess that was his choice.
Grading the exams took again around 9 hours, which is ridiculous. Last term I only graded like one question because there were a ton of exams and that one question took so long that I got totally burned out. This time I flew through the questions, I graded like five of them and just zipped right through, that was fun, made me feel productive. We also ordered this really spicy chinese food from some place that I guess was Szechuan, and Prof Meyer told us some story about how when he first dated his wife he was impressed by how well she kept up with spicy food. He also said she used to hate him because he was on her thesis oral panel or whatever and gave her a lot of shit because he didn't like the project that her supervisor gave. It's weird to think of him ever being young, he's just so old... That food was really spicy. Really really spicy...
Grades meeting was only 3 hours this time, thank goodness. Jay had a cool projected trends in miniquizzes and hw over the semester for every student, so that was helpful. The boundaries were also pretty self defining this time, which helped. I was happy almost all of my students did pretty well, so I got a good bunch this time.
That's that for that class. I'll probably have to TA again next semester, but hopefully Jay will be there still, forever one-up-ing me in TA semesters so that I never have to take on the head TA role, ever. =P