Skip to content
View MariusGen1's full-sized avatar

Block or report MariusGen1

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this userโ€™s behavior. Learn more about reporting abuse.

Report abuse
MariusGen1/README.md

Marius Genton - About me

I'm a junior at UCLA pursuing a B.S. in Computer Science & Engineering, with a technical breadth focus in Finance and Technology Management. I'm passionate about creating projects that align with my interests. Here, you'll find a showcase of some of my personal projects, including mobile apps, trading bots, quantitative analysis tools, and machine learning applications :)

Projects

BruinDining

~20k daily active users | 2M+ app sessions since launch

BruinDining is an app I built to facilitate dining for UCLA students, by making it easy to find open dining halls and their menus. Over time, I've added a lot of other features, including a rating & review system, notifications for favorite foods, and a meal swipe marketplace.

๐Ÿ“„ EEG Data Classification using hybrid CNN-RNN architectures

This paper's aim was to explore which deep learning architectures would be best fit to accurately map EEG signals from the brain to their corresponding muscle contractions.

Abstract

In this experiment, we analyzed 2115 sets of EEG data from 9 different subjects and performed classification among 4 classes. With the main goal being to determine the type of trial with high accuracy, we tried different archi- tectures based on conceptual reasoning as well as trial and error. We found that a model composed of convolutional layers performed the best, achieving validation accuracies of up to 72%. Furthermore, we determined that it is possible to get acceptable predictions after 500 time units out of the 1000 provided and that results from one subject did not seem to generalize well to the other subjects.

๐Ÿ“„ Read the full paper here

๐Ÿ’ป Jupyter Notebook available here

This paper was written for UCLA's ECE C147 class offering in Winter 2024 (Neural Networks & Deep Learning).

Aqua

โš ๏ธ Discontinued due to API shutdown

Aqua is a mobile app that displays data from Norwegian swimming meets in an easily understandable format. It also offers a multitude of features that go hand in hand with your needs while attending a swim meet such as qualification times for important events, a FINA-points calculator, and access to historic meet data. Full project here.

Climate Challenges (available on TestFlight)

Climate challenges is an iOS app that was ceated with the goal of encouraging people to act in a more environmentally responsible way. By offering a set of tasks associated with a points rewards, users can compete against each other to achieve the highest amount of points. In addition to making people act responsibly, the app is designed to educate people about how they can make an impact on the environment through simple changes in their day-to-day life. Full project here. Link to TestFlight download

Fly Away

Fly Away is a fast-paced iOS game that requires you to be on your toes, constantly ready to react quickly to unexpected situations. Your goal is to pilot the rocket to avoid the falling meteors, while also trying to get as many coins as possible in order to purchase upgrades in the lab. Full project here.

Cryptocurrency tool

This Python program is a tool that allows to test cryptocurrency investment stategies against historical data to determine their potential efficiency in the future. It allows the user to use a variety of different buying and selling strategies and provides detailed information about their performance. Full project here.

Binance Bot (Python)

This is a fully autonomous bot written in Python that invests a set amount in a cryptocurrency as soon as it is made available on Binance (usually within the first 1-2 seconds). Full project here.

Python Twitter Crypto Bot (Python)

Full project here.

Workout analysis tool

This is a workout anaysis tool built in Python that takes a CSV file containing workout data and magically outputs a understandable Excel file with useful graphs. The tool is built specifically to work well with TrainingPeaks' "export data" function. This way athletes can get an overview of their training over any selected period in an instant. Full project here.

Sample result

Pinned Loading

  1. Aqua Aqua Public

    1

  2. Binance-Bot Binance-Bot Public

    Python bot that automatically invests in newly created cryptocurrencies via the Binance API

    Python 2

  3. Twitter-Crypto-Bot Twitter-Crypto-Bot Public

    Crypto trading bot built in Python that trades based on Twitter events.

    Python 3

  4. Stocks-backtesting Stocks-backtesting Public

    A tool that allows to check whether certain data can be used to predict a price evolution

    Python 1

  5. TrainingPeaks-Analysis-Tool TrainingPeaks-Analysis-Tool Public

    Python analysis tool for converting TrainingPeaks data (csv format) to a visually intuitive Excel file

    Python 3