As a child, the Nintendo DS awoke something in me. It was the single greatest piece of technology I had ever owned up to that point (and probably ever will be, to be perfectly honest). I once played it for so long in a single sitting that my hands actually turned blue. Suffice to say, I loved my Nintendo DS. And of all the games on the console, I played Super Mario 64 DS the most. And of all the minigames on Super Mario 64 DS, I played “Wanted!” the most.
Two weeks ago, I watched a YouTube video of someone revisiting the game in 2022 and playing through the different minigames offered, and was reminded of “Wanted!” again, over 10 years later. As I watched the video, the wheels began to turn in my brain.
After so long of feeling burnt out as a data scientist, I finally felt inspired again!
I know this is not exactly the most technically challenging side project in the world, but I feel like the best thing to do for me in this situation is to value a project that gets me excited to code over something one that is potentially more impressive. And at the end of this work, two weeks later, I can confidently say I had fun working on it 🙂
I’ve published this blog on Medium, but I’ve disabled ads on the article, so you should be able to read it on there for free.
Letting an object detection model control a DS emulator to become an expert in playing the Super Mario 64 DS minigame, “Wanted!”
All the code used for this project can be found on my GitHub here.