Sunday, July 31, 2011

I guess it's a good sign that I'm actually wanting to investigate making minor changes to the models I'm working with, because I'm, you know, a RESEARCH INTERN

July 31: A Rather Typical Day

This morning I got up feeling rather nice, as I was able to sleep soundly until just before my alarm went off.  You can't ask for better timing than waking up naturally right when you would do so anyways. >_<  I think tomorrow I am going to try for a run, if I can will myself out of bed on time that is.

Today's dorm breakfast was chikuwa with cucumber along with some natto (and rice and miso soup).  Sigh... I guess I finally have to admit that I actually like natto now... D:  There's no denying it; I was happy when I saw it, and I now associate the smell with eating something I like, which means I don't mind it so much anymore.  Hopefully the next time I get a whiff of actual stinky feet I will not get hungry.  That would be more than a little disturbing. D:<

I got to work on time and wrote my first test program for my recently-completed recommender.  Amazingly, once I got everything to stop segmentation-faulting (an error all CPSC students are intimately familiar with), it just kind of... worked. >_<  I've only tested the recommender with 10 latent topics for faster feedback, but the prediction accuracy is kind of in line with what I'd expect...  It's kind of wierd when stuff actually just works.  I fiddled a bit more with the initial test program I wrote, then moved on to making the full-on test program that I'll use to generate results for a future presentation.

At lunch I ate Murugi curry, because the sobas and ramens didn't look very interesting today.  Murugi curry is supposed to be Indian or something, but really it's just Japanese curry with some stringy chicken in it. >_>  I talked pretty much all of lunch with Nishida-san about Evo haha.  I'm quite happy to have another fighting game nerd in such close proximity!  He was able to find match videos on NicoNico last night, which I don't have an account for, and I was only able to find ones on YouTube once I got home today.  Nishida-san wasn't familiar with North American players prior to this Evo except for Justin Wong, but I've been telling him about some, and he's been watching videos.  He's excited that some of the NA players play low-tier characters, as according to him, far too many of the top Japanese players just stick to Yun/Yang/Fei Long.  He quite liked Juicebox Abel's stage antics, and I am excited to recommend the Alex Valle vs. Daigo Umehara match from the ro32 tomorrow, if he hasn't already found it by himself.

I played handiball again today, and I found a better way to serve I think.  I found that I had to be really careful with my old way or I would crack my shoulder, but the new way seems to let me really smack the ball without fear of retribution from my joints.  People again passed to me today, so I guess they really were just not listening to me when I said everything except for serving was fine. >_<

In the afternoon I worked more on my recommender and the associated test code, along with reading some more about the other recommender I'm going to implement in the near future.  My next set of tests revealed another segmentation-fault-generating bug, which was at least simple to track down, if a bit drawn out.  At the moment (and I will eat my words some day, possibly soon), I much prefer it when my programs fail spectacularly, because that generally means that tracking down the point of failure will be obvious, as opposed to them thinking they're doing fine, but ending up with like 0.1% recommendation accuracy.  I guess that's what they taught us in class though, logical errors are much more obnoxious in practice than actual exceptions.

I left work at 6:15 due to another amazingly luckily-timed finishing of a part of my test program.  I got to the point where I could start a batch of tests on one of the variables I want to look at in my model tonight, so I set it up and then left the office.  I walked home again, and it was spitting rain now and again, but nothing too serious.  It is already getting a little dark at 6:30-ish here, which is amazing.  We don't have daylight savings, but I think it's probably still light out at 7:30 in Canada, isn't it?  It must be, or it would be easier to get Mom to go for bio-luminescence swims up at the shack (I'm just joking Mom, you are a very good sport).

Dinner was curry-udon, which was alright.  It was edible and filling, but I am finding that I prefer not to eat the foods I am very familiar with, because then I can see pick out things that are wrong with it haha.  If I eat something totally new to me, then I can't compare it to rather less-student-budget-feasible versions I've had in the past (eg.  sake and several pounds sukiyaki beef in the curry, etc etc).  >_<  Oh well, it was good enough and had lots of carrots and mushrooms and stuff in it.  There were also some squash bricks on the side that were incredibly dry by themselves, but quite good when dipped in the curry-udon roux. :D

After dinner I scurried up to my room to check on Evo and MLG results, and found Evo ro32 and ro16 matches on YouTube, much to the misfortune of the rest of my evening.  The other unfortunate thing about watching Evo matches is I really get the munchies, and end up eating a grapefruit and an entire bag of senbei in the space of 1.5 hours. >_<  Unfortunately, I think every-day grapefruiting has to stop, as I noticed my teeth were MUCH more sensitive today, even before the grapefruit.  If they continue to be a (literal) pain, then I guess I can go back to eating the grapefruit, but honestly I would rather have non-sensitive teeth. D:

Anyways, now it's time for bed so I can hopefully get up at an inhuman hour to run up hills and make myself want to be sick!  Hurrah... >_<

Night!

No comments:

Post a Comment