### Status: 🟢 Online

**Made with Angular**

## [View the site](https://yukseltron.github.io/Josephus-Problem/)

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2F3I0GxLCNCCsugnmNQs4m%2F1971257a-068e-4dd3-8c25-bd4190540be4.gif?alt=media&token=e2d0ddd5-36bc-4e14-a1dc-d3aa83a8e354","id":"1971257a-068e-4dd3-8c25-bd4190540be4","width":1316,"height":1190,"filename":"jp.gif","type":"image/gif","caption":"","border":false},{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2F3I0GxLCNCCsugnmNQs4m%2Fe83da13e-21e0-4593-8b62-42a1ec2932df.gif?alt=media&token=8332bda1-1d23-45cd-a783-b6666df034d3","id":"e83da13e-21e0-4593-8b62-42a1ec2932df","width":600,"height":411,"filename":"3.gif","type":"image/gif","caption":"","border":false}]}
```

This is a simple visualizer of the Josephus Problem. This was really just me trying out [Angular animations](https://angular.io/guide/animations). Part of the trick here was creating an evenly spread out circle of circles that could adjust to various sizes.

The Josephus Problem

**Made with Python and Pillow (Python Imaging Library)**

## [View repo](https://github.com/yukseltron/scribbles-art-game)

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2F8EOowVYe1j2AMoZ2polV%2Fcee01bf2-895a-495a-a6c8-791d1307ded0.gif?alt=media&token=965a2253-f0b6-473a-b3a7-6d40fae4ad3d","id":"cee01bf2-895a-495a-a6c8-791d1307ded0","width":250,"height":250,"filename":"1.gif","type":"image/gif","caption":"","border":false},{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2F8EOowVYe1j2AMoZ2polV%2Fb72c5f61-4125-4409-a357-7897d644c80a.gif?alt=media&token=69eb8c57-55dd-466b-92a8-3471ebfde75b","id":"b72c5f61-4125-4409-a357-7897d644c80a","width":1384,"height":794,"filename":"2.gif","type":"image/gif","caption":"","border":false}]}
```

This was a fun foray into image creation through code. Python's PIL is amazing at manipulating images, and the Pillow fork allows it to be even easier to implement.

As well, [I was inspired by this project](https://www.freecodecamp.org/news/how-to-create-generative-art-in-less-than-100-lines-of-code-d37f379859f/), in which pixel "video game enemies" are randomly generated. Of course, I wanted the images to look much different, so I took the approach of creating "scribbles" through the use of random shape outlines and vectors overlapping each other with random gradient colours. Doing this again with the same colour would create a second similar image to then create the GIFs you see above. \
\
Extending off of this, it was then added to a simple "shotgun" game to truly make it a video game boss, complete with a randomly generated name and quote.

Generative Scribbles Art

### Status: 🔴 Offline

**Made with Python Django,  IBM Watson Tone Analyzer, Aztro astrology API, originally hosted on Google Cloud.**

## [View the repo](https://github.com/yukseltron/emoji-horoscope)

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2FJuL71zWht8StqLsaz2S7%2Fc4275213-9f02-4178-a092-e7d109cfc606.png?alt=media&token=e373f9da-be35-44b7-b296-c31623110db3","id":"c4275213-9f02-4178-a092-e7d109cfc606","width":1172,"height":516,"filename":"Screen Shot 2022-01-22 at 10.03.07 PM.png","type":"image/png","caption":"","border":false},{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2FJuL71zWht8StqLsaz2S7%2F8b47d0ab-2170-4ced-bfbb-be6c4084b4e3.png?alt=media&token=a576f57c-16cd-45b4-9e2a-11230668ae80","id":"8b47d0ab-2170-4ced-bfbb-be6c4084b4e3","width":1022,"height":1034,"filename":"Screen Shot 2022-01-22 at 10.03.27 PM.png","type":"image/png","caption":"","border":false}]}
```

The focus for this project was on trying out Django (a Python web framework) and two API's. The web app is pretty simple, users input their astrological sign and then this makes a call to the Aztro API for the text horoscope. This text is then fed to the IBM Watson Tone Analyzer, which I then map the results of to specific emojis.

```
joy = ['😁','😀','😂','😆','🤑','🤣','😊','😋','😍','😇'] anger = ['😡','🤬','😠','🤯','😈','😤']
fear = ['😱','😰','😨','😬','😳','😥']
sadness = ['☹️','😖','😞','😢','😭','😕','🙃','😔']
analytical = ['🤔','🤨','🧐']
confident = ['😛','😜','😏']
tentative = ['😒','😑','😓','🤭']
```

In reflection, I could definitely work on improving the mapping from tone to emoji. I think maybe it could be more nuanced. But this also depends on the list of tones IBM Watson provides. Of course, I could also explore other sentiment analysis libraries.

Emoji Horoscope

### Status: 🔴 Offline

**Made with Discord.js (a Node.js module for the Discord API) and TheMovieDB API.**

## [View the repo ](https://github.com/yukseltron/buscemi-bot/tree/master)

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2FYRiaJI2aw8MJvyNBYXev%2F4784a401-b757-4480-830f-3163cca991c2.png?alt=media&token=928a55cf-fc79-4f0e-8c04-eff7697893f0","id":"4784a401-b757-4480-830f-3163cca991c2","width":1368,"height":498,"filename":"Screen Shot 2022-01-22 at 1.12.58 AM.png","type":"image/png","caption":"The unfortunate takedown request :(","border":false},{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2FYRiaJI2aw8MJvyNBYXev%2Feec16230-8f03-40b3-bd36-ade8d2550e70.png?alt=media&token=25ae7e9f-e356-49b3-b5c3-476351c5501b","id":"eec16230-8f03-40b3-bd36-ade8d2550e70","width":1120,"height":1092,"filename":"Screen Shot 2022-01-22 at 1.09.30 AM.png","type":"image/png","caption":"","border":false},{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2FYRiaJI2aw8MJvyNBYXev%2Fb4316759-1de2-4887-b387-ec3fbe166e0f.png?alt=media&token=58fe80fb-7ea5-473e-bd32-cb6aaf8ce08d","id":"b4316759-1de2-4887-b387-ec3fbe166e0f","width":1130,"height":584,"filename":"Screen Shot 2022-01-22 at 1.11.26 AM.png","type":"image/png","caption":"","border":false}]}
```

This project was me trying to make a discord bot. Using TheMovieDB API, I created a bot that emulates "Steve Buscemi". This bot would be able to fetch quotes, clips, films, and images of Steve Buscemi. Why Steve Buscemi? I'm not entirely sure why. Looking back the project seems ridiculous. In actuality, the bot can be adjusted to act as any actor listed on API, I just for some reason chose to make it Steve Buscemi as an example. Naively feeling proud of the project, I even submitted it to [Discord Bot List](https://discordbotlist.com/). This way, the Buscemi bot would be hosted on an external server. \
\
However, before I could get the bot listed, I received a takedown notice from TheMovieDB. It had been sent from a certain someone I can only assume is associated with Steve Buscemi. Somehow, they had discovered I was using his likeness. I was surprised at the time, though I guess I simply did not understand the legal aspects. I had thought because I was not trying to monetize anything, it'd be ok. Oh well. I hope Steve Buscemi does not hold negative feelings against me for this.

Buscemi Bot

### Status: 🟢 Online

**Made with Knockout and the Giphy API**

## [View the site](https://yukseltron.github.io/giphy-tv/)

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2Fduqstmp2sy2fmQSVqY3o%2F4c57d322-dc53-414d-a8e9-6493c51dafed.png?alt=media&token=5ac59b2b-e00a-4227-b811-99e23d468b4e","id":"4c57d322-dc53-414d-a8e9-6493c51dafed","width":1972,"height":1304,"filename":"Screen Shot 2022-01-22 at 11.04.18 AM.png","type":"image/png","caption":"","border":false},{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2Fduqstmp2sy2fmQSVqY3o%2F5656b072-f66e-4bfd-bc45-f4b670fc8a65.png?alt=media&token=3994ec9c-239c-44bf-bfb6-4b35c33c1fea","id":"5656b072-f66e-4bfd-bc45-f4b670fc8a65","width":1972,"height":1304,"filename":"Screen Shot 2022-01-22 at 11.04.40 AM.png","type":"image/png","caption":"","border":false},{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2Fduqstmp2sy2fmQSVqY3o%2Fad5c6b7c-56f7-4f61-8046-89bbf90c985b.png?alt=media&token=4272b285-4fd7-4f45-8f3f-258d493d283f","id":"ad5c6b7c-56f7-4f61-8046-89bbf90c985b","width":1972,"height":1304,"filename":"Screen Shot 2022-01-22 at 11.05.15 AM.png","type":"image/png","caption":"","border":false},{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2Fduqstmp2sy2fmQSVqY3o%2F45740ddb-e0cc-4a40-acf8-dd8620f90e55.png?alt=media&token=e758bda2-ee77-4e49-9d7e-3d480600e1d1","id":"45740ddb-e0cc-4a40-acf8-dd8620f90e55","width":1972,"height":1304,"filename":"Screen Shot 2022-01-22 at 11.04.14 AM.png","type":"image/png","caption":"","border":false}]}
```

An early project of mine back when I was still new to web dev. At that time, all these new JS frameworks and libraries were all the craze. Ember, Meteor, Angular, Knockout, Backbone, Vue, React, it was fun going through them and trying them out. Though things have changed and now I'd probably just stick with using React. This is the only time I've used Knockout. I had heard about it from a hackathon I had been to, and a friend recommended I try it out.  At the same time, I had been interested in exploring the Giphy API, so things sorta fell into place as I just tried making something simple that combines both of them.

GiphyTV

**Made with Python and CatBoost**

## [View notebook](https://github.com/yukseltron/ML-lyric-classifier/blob/master/2179-proj.ipynb)

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2Fg5Dm12Yk91heDfXkyTtp%2Faad3f923-fa3e-4d2a-9723-2e226fd29270.png?alt=media&token=477f79d7-3222-4b11-9a23-fcdb14533530","id":"aad3f923-fa3e-4d2a-9723-2e226fd29270","width":1312,"height":1122,"filename":"Screen Shot 2022-01-23 at 8.24.11 PM.png","type":"image/png","caption":"","border":false}]}
```

This project was for a machine learning applications course. It was completed within a week and was the first ML model I've worked on alone. The project goal was the same for everyone: given a dataset of real music lyrics, classify them into one of three genres (Rock, Hip Hop, and Pop).

### The Story

I needed to build a classifier, but I lacked domain knowledge in lyric composition. While there may be beautiful patterns that help sort the lyrics into a genre, I do not know any of that. So, it was hard to do any useful feature extraction from the lyrics. Things like lyrics length and number of verses vary a lot so simply finding quantities would not be enough. With limited time to work on this, it'd be difficult to gain a deeper knowledge of lyrical composition.\
\
The challenge here was in processing the lyrics. It was a (long) text feature, and so as a categorical data type is a bit tricky to convert into a numerical feature for use in other regression methods. That's when I began to wonder if I really needed to work that hard in converting the categorical data to numerical. Surely, there must be something already out there to handle text features easily.

While exploring some ML libraries, I came across CatBoost. CatBoost employs gradient descent and was especially interesting for its focus on working with categorical data. A little digging deeper into the CatBoost docs (which was also a challenge), I found that CatBoost actually has support for text features. \
\
Implementing CatBoost on the lyrics was then super easy:

```
#Build the model
model = CatBoostClassifier(iterations=100,
                           learning_rate=3,
                           depth=6,
                           loss_function='MultiClass')
```

### The model's accuracy: \~68-70%

This accuracy turned out to be really good, comparatively speaking. But, I didn't realize how good when I submitted it. In fact, I had fumbled my first submission, using testing the model on the holdout set rather than the testing set to my embarrassment. So when I submitted this for a second time at 3 AM on the day before the deadline, I was really just hoping I at least had done ok.

### Draft 1 accuracy:

> **From:** ML Professor
>
> The ABSOLUTE distance between your predicted accuracy and the real accuracy is: 0.125\
> \
> This is quite a lot.

When I submitted my first draft for testing, it turns out I had made quite some mistakes. CatBoost has extensive docs but lacks in greater details of its capabilities. This can sometimes make it hard to understand what CatBoost can do and how. Regardless, after the poor results of the first submission, I had to decide on what my next approach would be.\
\
CatBoost was not explicitly taught in the course. The professor encouraged us to use libraries outside those he had demonstrated in class, so I was immediately eager to find something new. I don't think I could carefully explain to you how gradient boosted descent works, but nor did I need to for the context of the project. But still, here I was staring at my broken CatBoost model, wondering if I should throw it all out the window and use something more tried and true like K-Fold and Naive Bayes. Instead, I chose to keep going with CatBoost. I believed if I could research it more, I'd understand how to properly implement the text feature.

### Draft 2 accuracy:

> **From:** ML Professor
>
> The ABSOLUTE distance between your predicted accuracy and the real accuracy is: 0.001
>
> You have a very good accuracy, congratulations!

For the second draft, I looked further into CatBoost and outside the docs.  After watching [this very helpful video](https://www.youtube.com/watch?v=usdEWSDisS0&t=1s), I was able to find and fix the mistakes I had made prior. Checking the accuracy of the model, this time it looked much more promising! 

After submitting it for the final time, I later found out during a class that the submission with the highest accuracy was one that used CatBoost. There was also only one person apparently who used CatBoost, so I guess we can infer whose submission that was ;)

Lyrics Classifier

### Status: 🟢 Online

**Made with Vue.js and hosted on Vercel**

## [View the site](https://monodrift.vercel.app/)

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2Fgyxy8sMVmfwSUJKJADOy%2Fd66e4267-3884-456a-956b-4b5577711792.png?alt=media&token=08e3006a-013f-4fc7-b0d2-f5da01f0dab9","id":"d66e4267-3884-456a-956b-4b5577711792","width":2880,"height":1330,"filename":"Screen Shot 2022-01-21 at 11.59.44 PM.png","type":"image/png","caption":"","border":false},{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2Fgyxy8sMVmfwSUJKJADOy%2Fbaccf2ca-7615-4709-b359-fcadb0da082d.png?alt=media&token=561b6c97-2af0-4821-955d-e8ce39df00ba","id":"baccf2ca-7615-4709-b359-fcadb0da082d","width":2874,"height":1328,"filename":"Screen Shot 2022-01-22 at 12.04.21 AM.png","type":"image/png","caption":"","border":false}]}
```

I'm a big sci-fi fan. And sometimes, that ends up with me trying to do some writing. Of course, I am not a talented writer and have little experience writing in long-form. So, I just wrote these super-short stories with a sci-fi flair to them. Apparently, I later learned from a friend this form of writing could be considered as [Flash Fiction](https://en.wikipedia.org/wiki/Flash_fiction), where the word count lies at around \~1000.

Anyhow, this project also involves a development side. I wanted to work on a small Vue.js side-project while also putting in some writing. The website was easy enough to make, the content? Not so much. Since I've only written a few of the stories (which looking back, I do think I could do better now), there are some that are unfinished.

I do have notes that I've written somewhere in some journal that details what I want the other stories to be about. So one day, I think I will come back to this and rewrite it with new lenses. Literally too. My glasses are new.\
\
With a whirr, the machine spits back out a small grey card.\
The card is chipped and bent at the corners.\
Parts of it still retain its gloss, while most of it is skinned, exposing thin strands of plastic bleached by the sun pealing off.

Mono-Drift

### Status: 🔴 Offline

**Made with Swift**

## [View the repo ](https://github.com/yukseltron/buscemi-bot/tree/master)

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2FvbS24hYQEyztEcR626ti%2Fe3057f79-a321-4850-b242-5308d30e886b.png?alt=media&token=d83a2790-6b42-47e7-b6cf-d939c4be349c","id":"e3057f79-a321-4850-b242-5308d30e886b","width":1168,"height":2096,"filename":"1.png","type":"image/png","caption":"","border":false},{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2FvbS24hYQEyztEcR626ti%2Ffa203471-0858-40de-a0a8-33cb581ee249.png?alt=media&token=4b275a08-3830-48a4-885c-acf1294e30c8","id":"fa203471-0858-40de-a0a8-33cb581ee249","width":1168,"height":2096,"filename":"2.png","type":"image/png","caption":"","border":false}]}
```

This project was part of a university class working with iOS development. The goal was to build an app for helping you journal and track your house plants. While this is a common app, the difference here is that this app focuses not just on time since watered, but more importantly on the moisture of the dirt. \
\
After reading a very interesting book called *[The New Plant Parent](https://www.houseplantjournal.com/product-custom/the-new-plant-parent/),* it was interesting to learn the importance of maintaining different moisture levels for different types of plants. This, combined with the light level the plant receives, is far more important than simply rewatering it on a consistent basis. \
\
As such, this app would have a registry of common houseplants and provide the appropriate level of moisture and light the plant prefers. As well, a calendar was also implemented for tracking water days. 

\
While this app is not on the iOS store, it was a fun introduction to Swift but a brutal introduction to SwiftUI. SwiftUI had just come out around the time I started this project and I went ahead with trying to implement it. It was a bit tricky, to say the least. A newer tool lacking in appropriate documentation at the time, SwiftUI provided a challenge to implement. But... I can definitely see the potential in SwiftUI. I think in theory it does make the UI much easier to implement than before, so I do look forward to seeing it develop and mature.

```cv
gallery
{"layout":"hscroll","images":[{"src":"https://firebasestorage.googleapis.com/v0/b/maitake-project.appspot.com/o/pages%2F6KQG3gmhSra20VnXw3yALJ7M20J3%2FvbS24hYQEyztEcR626ti%2Fed67bd65-03c3-49ad-8738-3a121922ada8.jpg?alt=media&token=ac3bd93f-61a4-4a3d-ac4d-f4dc481cd747","id":"ed67bd65-03c3-49ad-8738-3a121922ada8","width":1149,"height":1532,"filename":"65679253891__CD89D953-C3FB-4D0E-9B87-EFD7461BA4E3.jpeg","type":"image/jpeg","caption":"The secrets are in the soil","border":false}]}
```