Past Projects:
Web Racing Game - Backend - Data Engineering
β Tools: Redis-json, Nodejs, Express, JavaScript
β Built backend and server logic for web app racing game using in memory database and live data streaming.
β Won gold prize ($2000) in Redis 2021 Hackathon. π₯
Malaria Blood Smear Classifier - Image Classification
β Tools: Python, TensorFlow, TensorBoard, Matplotlib, Pandas, Numpy
β Used a public dataset of 23,000 images to classify with 95% validation accuracy
β Identified and visualised mislabeled data to quantify validation error
NBA Betting Model Project - Regression Analysis
β Tools: Python, Numpy, Scrapy, Pandas, Sklearn, XGBoost, Matplotlib, Seaborn
β Predicted the total score of a match within 15% accuracy of the betting lines
β Scraped, cleaned, feature engineered and hyper-optimized for improving model results
Business Expansion Project - Geo-spatial Interactive Visualization
β Tools: Python, Pandas, Geopandas, Contextily, Folium, Matplotlib
β Located and visualized high population area of the city, with low business competition
β Analysed Google ratings to find most favourable gym type for highest success
2016 Canadian Census Visualizations - Exploratory Data Analysis
β Tools: Python, Pandas, Matplotlib, Seaborn, Squarify
β Read and sorted census data to provide insights with various visuals
β Visuals used: Vertical/horizontal bar graphs, pie-charts, box-plots, and line graphs
Model Selection Guide - Beginners Guide
β Summary of machine learning models commonly used for regression / classification (from scikit-learn) β Reading through documentation about these models can often be very intimidating for beginners, especially when you do not have a strong math background. This is meant to be written for more people to understand the difference between models.