Hi, I’m Sabri Snyder and I’m part of the Programming and Robotics 8/9 Class. Below are some projects I’ve done at home and at school to showcase my skills learned in class.
My OBS YouTube Video Link:
NHL Point Predictions
We looked at API’s and used past player’s performances to predict their future performance. Ultimately, from mid February to mid April, I was off by pretty much exactly 1 goal from my expected amount of goals from my 25 players, which I think is pretty impressive and shows the potential in such a crude strategy for calculating player performances.
<iframe width="481" height="298" seamless frameborder="0" scrolling="no" src="https://docs.google.com/spreadsheets/d/e/2PACX-1vT6MGT_HKoR4ZaiK2MrV35DdikOh6LiuHe8tjhphmM3EVpkRUh7G1oTfyc289wH29lmJy5B6FXgrrQf/pubchart?oid=1737461343&format=interactive"></iframe>My trinket coding:
Rock, Paper, Scissors
<iframe src="https://trinket.io/embed/python/2bb32e10ff" width="100%" height="600" frameborder="0" marginwidth="0" marginheight="0" allowfullscreen></iframe>Age Calculator
<iframe src="https://trinket.io/embed/python/8a3f3396e7" width="100%" height="600" frameborder="0" marginwidth="0" marginheight="0" allowfullscreen></iframe>Simple Timer
<iframe src="https://trinket.io/embed/python/d626b282b0" width="100%" height="600" frameborder="0" marginwidth="0" marginheight="0" allowfullscreen></iframe>Number guessing game, 1-100
<iframe src="https://trinket.io/embed/python/cdb7685430" width="100%" height="600" frameborder="0" marginwidth="0" marginheight="0" allowfullscreen></iframe>
1st Commentary on Tech News: ChatGPT4-o
OpenAI recently made GPT-4o available to everybody who has an OpenAI account, and GPT-4o has some very scary and very innovative features. The first one I’ll talk about is the ability to have a “real” conversation. GPT-4o has a voice and can actively participate in a conversation with 1 or more people. The voice and the mannerisms of GPT-4o are already way ahead of every other AI voice. The voice sounds very close to being real, and the AI knows how to make jokes, laugh, and respond at correct times. Although it isn’t perfect, and sometimes struggles to understand the context of a conversation, most of the time it’s pretty accurate. What makes it even more realistic is the response time. The average response time of GPT-4o is 320 milliseconds, and it can respond as quick as 232 milliseconds. In an interactive human conversation, the average response time is around 250 milliseconds. This means that GPT-4o is nearing human response speed, sounds like a human, and can communicate like a human. Less than 2 years ago, ChatGPT was launched, and since then, AI has grown at extremely quick rates. The possibilities in the near future for AI look very innovative and interesting. Personally, I think that it may be a lot sooner than we think before AI begins to replace people’s jobs, because already people are trying to include AI into robots, and once that becomes publicly available, assuming these robots could function well, many jobs could get replaced causing a job shortage worse than what we saw from the 2020 Coronavirus pandemic.
2nd Commentary on Tech News: Apple Vision Pro
The Apple Vision Pro was released on February 2nd to the United States and still has yet to release globally to the rest of the world. The Apple Vision Pro is a headset with both Virtual Reality (VR) and Augmented Reality (AR) into one. It makes for an immersive experience, it’s good for entertainment, and regular productivity similar to that you could do on an iPhone. It’s very high quality, but right now gets dwarfed by other headsets on the market due to similar quality to some Meta headsets while being significantly more expensive. Apple definitely has plans to make better versions in the future, and the first ever Apple Vision Pro wasn’t really meant to be sold to the public; at least it hasn’t been optimized to do so, which will change in years to come.
Personally, I will not even think of purchasing the Apple Vision Pro in this form. The only reason I would ever purchase a headset as of now is for gaming, and Meta currently dominates that market, so I would rather take a Meta Quest headset. Now, although it has some unique features, it doesn’t outweigh the fact that it is around 3x the price of the Meta Quest in my opinion. I believe what Apple wants is for this to be a mainstream thing that everyone will own for their every day life. If so, then this is the correct direction, or at least that’s what it seems like to me. I think that if it gets lighter, easier to use, and cheaper, it could become as common as an iPhone, although I think it will be a little more alien and will take longer to be integrated into society rather than the iPhone which took over the world in less than decade.
Testing an AI checker: Comparing my Essay to An AI Generated Essay to see if an “AI checker” can spot the difference
After inputting my essay into GPTZero, it gave my essay a 7% chance of being written by AI. I, of course, wrote my essay entirely without the aid of others, or AI. Then, after giving the AI the same prompt I had for my english essay, and then copy and pasting that into GPTZero, it gave that essay a 100% chance of being written by AI. Pretty remarkably accurate, even if it is a small sample size. Then, after asking ChatGPT to write this essay again, this time to go undetected, GPTZero came back with a 100% AI chances again. Although things changed when I ran the AI generated essay through Quillbot. Quillbot is a paraphrasing AI that can be used to try and go undetected by AI checkers such as GPTZero, and these results surprised me. After running the essay through Quillbot.com, GPTZero said there was only a 17% likelihood of the essay being written by AI. Ultimately, a very interesting experiment to test the accuracy of AI checkers.