I’m writing about my experience planning this recital and making it a reality two days ago, on 4/30. For my final recital, I wanted to perform a few of the pieces that I learned and practiced throughout the senior project. I chose to play Bach’s Ave Maria, Saint Saens’ The Swan, Massenet’s Elegie, and Kreisler’s Liebesleid on the violin, and the complete Walton Viola Concerto (on the viola), and topped it off with a duet featuring my dear friend and violinist, Marco Opena, a student of BASIS Scottsdale and my stand partner in the Phoenix Youth Symphony.
Three weeks ago, I reached out to another music teacher at BASIS Mesa, Ms. Chiang, to ask if she could play the piano in my recital. After a planning process that took place over the span of two weeks, it seemed like our schedules would not work together for her to make my recital. We settled on an alternative, on the day of my recital, I would come to her classroom during a break and play together with her, for no particular audience, simply for the joy of making music. With a stroke of luck, we discovered that somehow she would have to stay later through the afternoon, and actually could come to my recital after all. So we had the most spontaneous performance of two pieces at my recital, having only rehearsed for 20 minutes moments before.
In contrast, my duet with Marco went much more smoothly. We chose to play Passacaglia by Handel/Halvorsen. We met a week before the performance for our first rehearsal and brought the piece to his lesson, with a violin professor at ASU. Through our rehearsal and lesson, we were able to prepare the showpiece sufficiently well to confidently finish off my recital.
Last week was my last week at Classic Promenade. I spent the day with the mechanic James, and said my goodbyes to the rest of the team. I talked with James about what I plan for my future and he wished me luck on my journey. I thanked Harry for giving me the incredible opportunities and experiences that will help me in the future. I spoke briefly with Harry’s wife Heather, who has been extremely kind to me, about how I have committed to the University of Arizona and she wishes the best for me in college.
I am extremely grateful for this internship and the help I have received along the way especially from Giles and Mr. Kirschner who have both supported me and pushed me to step out of my comfort zone and to try new things.
Looking back, this research assistant opportunity at ASU has been an incredibly valuable experience for me, and I’ve certainly learned a lot of new things. I remember going from needing help setting up my computer environment and installing python, to learning how to create my own basic machine learning models with Tensorflow.
In fact, for my final product, I decided to create a machine learning model for making general predictions about the stock market based on quarterly macroeconomic signals like GDP growth, interest rates, and unemployment. It turned out to only have about a 73% accuracy rate, but I suppose that’s to be expected.
I’m incredibly grateful for the support of the professor and the PhD students that I’ve worked with. They’ve made it possible for me to grow and learn so much in such a short period of time.
Although my senior project is coming to end, I plan on continuing with my internship for at least another few weeks. I hope to see through the end of the GAN project and help out a little more with the AJP project. I’m also still uncontent with how much I know about coding AI models, so I’ll continue with my own studying and practice with Tensorflow.
My next steps will be to expand upon my current knowledge in college and hopefully these types of skills in a career. Right now, the plan is to double major in Computer Science and Business, because machine learning algorithms can be extremely useful in the field of finance. Thanks to this experience, I’d say that I’ve already gotten a pretty good head start, but I don’t expect the next few years to be a breeze either.
I’m looking forward to seeing what the future will bring. AI has had some amazing advancements just this decade, so who knows how it’ll transform the world later on.
This past Thursday I performed my one man show. It was a runaway success. Everyone that was there complimented me on my delivery skills, said that it felt relatable to their issues, and even suggested that I should go into public speaking. I feel incredibly validated and am incredibly thankful for everyone who showed up. I hope to have the recording available soon.
My performance
To everyone who’s followed my senior project for the past 10 weeks, I would like to say a few words. Thank you for following this incredible achievement for so long. I hope you all had success in whatever career path you wish to pursue, and I wish you all the best in your future endeavors. Thank you.
So how is a book constructed? At the most basic level it is simply gluing the pages with content (text block) into a cover made of a more durable material, a technique often seen in paperback books. This carries a slight downside in the fact that it is terribly fragile but easily mended at least. More complex is to sew the text block together into segments before gluing it into the cover to provide more durability. This allows for another useful feature, the endband. This more effectively allows the usage of hard backed books as the endband increases the flexibility of the text block. Now with hardcover books some new elements come into play, a hard cover and endpages. The core of a hardcover in the modern day tends to simply be a thick paperboard (around ½ centimetre) which is then covered in cloth or a faux leather. The endpages are simply heavier pages, although on occasion a metal foil is used, pasted in between the covers and text block on each side and most always pasted directly onto the cover to cover the in facing paperboard. Common to all of these is the usage of book paste, which while not having super specific requirements needs to be PH neutral and flexible.
As for this project I’ll use the simplest method as there is no reason to do more. It will be bound with a basic cardstock cut down to scale as the cover with the text block simply glued in place.
Remember that ink project that I mentioned a month or so ago? Well, I’m back on it now. Like I said in my last post, the GAN project that I’ve assisting is slowly coming to an end, so there isn’t much left me to do anymore on that. As a result, I’ve went back to helping on another main project, the Aerosol Jet Printing one with the cool lines of ink.
The task I’m working on now is called semantic segmentation, which is basically just a fancy way of editing an image to help computer software better process visual information. Semantic segmentation is method of classifying and assigning labels to specific pixels within in an image. This helps deep learning models efficiently analyze the shapes and forms of objects and figure out where the important parts are. In order to do this, I have to segmentation masks, which are portions of the image that are differentiated from everything else.
A really simple way to do semantic segmentation is by just painting over the important parts of an image with a labeling software. This technique is sometimes pretty inefficient, so another way is by simply drawing in colored polygons over the image. That’s what I did in the image below. I put both the original image and the segmentation map version of it below. The red part is labeled as the central line, the green parts are the ink overspray, and the blue specks are the solvent overspray.
At this point in the college decision process, while my plans are still unclear, I realized that I will most likely study viola full-time and hopefully pursue a career on the viola. Ironically, I still can’t properly read sheet music written in alto clef (the viola clef), and I stand no chance if I needed to sightread anything on the viola. It doesn’t help that I play violin all the time for my senior project. I decided I needed to take baby steps to start getting more comfortable with the alto clef.
Yesterday I was playing in a corner near the cafe at Banner Baywood. I started with playing some nice violin stuff. About an hour in, I took a break and I wanted to find some things I could play on the viola for that occasion. To my surprise, I easily found the viola version of the Suzuki method online. I was already familiar with these pieces because they’re mostly the same as in the violin version, some even preserved all the same fingerings. I started reading some of these pieces, and the cafe waitress immediately noticed something, “Is that a different violin?” She evidently didn’t know what a viola was, but that’s alright, she said she really liked the sound of this instrument, “not that the other one wasn’t good.”
Later in the day, I was performing for an intimate audience at Inspira Gateway. I pushed myself to play the entire program on viola. I suddenly had an impulse to pull out the first movement of the Walton concerto, which I haven’t practiced for nearly two months. After finishing a sketchy first movement, I decided to play through the second movement, and then the third. Although it wasn’t great, I was proud that it was the first time I ever performed an entire concerto in one go.
So, illustrations are a lot harder on a canvas seventy-seven and a half times the scale you’re used to working on. Or more specifically the shading is, sketching is familiar enough as it’s already the base of drafting a sprite. I was hoping to use cell shading since it’s what is familiar but while the general process remains the same—choose mid-tone, highlight, shadow, and accent colours, then designate where each falls, and finally fill in each area—the actual process is different, requiring different tools and learning new skills (such as how to use a stabiliser). Line work is the other issue since that is something I have never had to deal with (though arguably its equivalent is deciding which shade and how many pixels a sprite’s outline is/if this part should be outlined). The general process is the same and this time I know what brushes to use and (generally) how to use them, but practice and muscle memory isn’t there. For reference, the general process is to first sketch it physically and scan it or just sketch it out digitally, then go over it with either a brush with static pressure or with pressure scaling size (typically a high curve since the line weigh should be mostly the same), then go and trace over sketch with as few strokes as possible (or at least have them connect). The last part is where I struggle making this simply a skill issue. So I’ll just finish up the required page (credits for font and other copyright stuff, though the font is the only one I think) and print and bind.
Last week was a bit different than my usual day at Classic Promenade. I had the opportunity to attend Radford Racing school for a tour and some hot laps on Thursday. I reached out to them a couple weeks ago telling them what I’ve been up to in regards to my senior project as well as my desire to pursue a career in motorsports. I was not expecting a response, however; they informed me that they would love to have me come down to the school to take a few videos to use for my final product. I brought Sarah along with me, because I knew she would enjoy this experience as well and so that she could help me record videos.
At 8 A.M. we showed up to the track and were welcomed by Christine who was interested to hear about my project. She introduced us to Danny who gave us a brief history and tour of the school. He drove us around in one of the Dodge cars and took us on a ride for some hot laps. That was easily one of the coolest things I have experienced, as we had to have been going at least 150mph. Everyone their was very welcoming and kind to allow us to wander the grounds and learn about what they do.
As I made last week’s post, a certain realization came to me that I was much closer to finishing the helmet than I anticipated. In just a few hours, I managed to get the whole thing assembled and officially complete:
This week I plan on doing more of a professional photoshoot of the helmet in more varying angles, but here is a look at the interior:
The crushed red velvet came along very nicely and bonded well with the closed-cell foam. There is about an inch of foam at the very top of the helmet and closer to a quarter of an inch on the sides. Overall, the helmet is very comfortable and doesn’t have any pressure points or wobble when it sits on my head. It is, however, quite heavy — the final weight, cloth included, is 4.8 pounds! That is roughly the same weight as a half-gallon of milk, but with the foam cushioning your head it isn’t completely overwhelming. It sure would have been much heavier had I made the helmet out of copper or brass, which are each around 10% denser than steel.
