Foreign Exchange Rate Deep Learning Project

Fascinated by the domain of forex rate forecasting, this project sets the stage for creating a model that captures the complex dynamics of the market. Neural networks and the A2C deep learning algorithm train an adaptive model that learns from experience and makes optimal decisions.

Gift Genie - LLM Chatbot

Gift Genie is a recommendation system that suggests gifts based on user input. Allows users to personalize gifts based on the recipient's age, gender, personality, and price range in US dollars. Used the Cohere LLM and hosted on Streamlit.

ResQLearn

Tackle the issue of the lack of emergency knowledge taught in our traditional education system. Designed a gamification of traditional emergency skills' education, ResQLearn, for children to learn vital rescue techniques in fun and innovative ways.

Transfer Learning of Chinese Characters

Exploring Traditional Chinese Optical Recognition. Focuses on classifying three commonly used Chinese characters: "的" (de), "不" (bu), and "和" (he). Leverages the power of transfer learning. Builds upon the knowledge acquired by a pre-trained model - ImageNet and ResNet.

Data Analysis and Visualisation of the Relationship among Wine Rating, Price, and Origin Country

This research analyzes a dataset of red wines, focusing on winery, average rating, price in EUR, and year of production. By exploring these variables, we aim to gain insights into the highest-rated red wines from year of production, ranging prices and wineries.

Predictive Analysis of Job Salaries in the AI, ML, and Data Science Space

This project aims to analyze anonymized salary information from professionals worldwide and develop a machine-learning model to predict salaries accurately to address the challenge of determining fair compensation based on one's experience level.

Q-Leaning Frozen Lake Game

This is a trained model of a Q-Learning agent playing FrozenLake-v1 .

Reinforcement Learning PPO Agent Playing with Lunar Lander

This is a trained model of a PPO agent playing LunarLander-v2 using the stable-baselines3 library.