Yusuf

I did not add Devlogs here for this project.
I am really sorry for that, though I did take some screenshots throughout building this and documented the process. (Which I will publish in a blog post, hopefully in a few days.)

To reiterate, this is a very basic proof-of-concept MVP. Here are more details on how I built it:

First of all, I tried creating my own CNN model by watching a bunch of tutorials, but failed (It’s easy to put in one sentence but it took me a LONG TIME and also a bunch of unlogged hours since most of it was on Google Colab)

Then, I realised I could use something called ‘Transfer Learning’ and researched for the best models, and I found ShuffleNetV2, which was just around 6 MB and had a lot of classes (Objects that it could detect), but I only needed 47 classes (Numbers + Capital Letters + Some Small letters, provided by the EMNIST dataset)

So I imported the model, ‘unfroze’ it and then re-trained it using Google Colab’s free GPU to create the model.

The next step was to build an OpenCV pipeline to preprocess the image. I already had some experience with OpenCV, so it was easy to start. On top of that, I watched a bunch of other implementations and tutorials on this to do this.

The accuracy is still poor, and I have many potential improvements, but I was able to get the model under 2 MB using the methods described above.

I converted the whole code to Webview using emscripten, and got a bunch of errors but finally figured it out and hosted the code on a website using Github Pages.
A few final tweak remain but otherwise I am ready to ship the first game in this library. I might actually host this on my personal blog, which is work-in-progress for now.

Not much to update, just tried to re-adjust file structure and it was literally brutal, errors after errors and now I have managed to get it working again…
I will now try to compile it to web (Which I tried a bit before, was a headache but yeah)

So I completely broke it, wanted to change some logic to add ‘sliders’, so had to change an array to a vector and even after changing all logic, it just kept breaking, after working an hour or two I got it working and then added a button and GUI.
(Should’ve made git commits, but I haven’t properly set up the repo for git, since it’s C++ and I coded my first line of C++ like 2 days ago)

Ok so I finally got the first simulation working:

The Chaos Game! Here are the rules:

1. 3 random points are randomly picked anywhere on the screen
2. A random point is picked completely randomly on the screen
3. The random point then randomly chooses any of the 3 points and then a new point is added halfway through it.

Using so many randoms, one would’ve guessed that it was a messy pattern but that’s not what happens in the simulation!

Now I will work on adding GUI, ability to change other parameters etc.

(Keep in mind I have NEVER coded in C++ before and this is my first time EVER)

Have to do a COMPLETE restart.

This was originally “Interactive Mathematics In Python” but because I was using streamlit with matplotlib, it just wasn’t efficient, I kept running into errors and gave up midway.

I am now using C++ to code, and would not be making a web app. All the best to me.

Screenshot is just the ‘home’ page in streamlit, when it was written in Python, I have scraped it now.

I’m working on my first project! This is so exciting. I can’t wait to share more updates as I build.