This project is to help Home Credit to make predictions about clients capability of repayment using various statistical and machine learning methods. Please see the report for details

Created several unsupervised model such as K-Means, Isolation Forest and OneClassSVM which flags suspicious traders to send for additional review. Won the Best in Presentation

Collaborated with team members to analyze rugby athletes’ wellness of being and competitive performance, where we applied Pandas for data cleansing/wrangling, constructed the visualizations and built XGBoost, neural network, K- Prototype clustering and regression models. Presented to 20 judges and hundreds of participants and won the Best Visualization

Fitted both generative and discriminative models to the MNIST dataset of handwritten numbers, where implementation has been vectorized.

Implemented a probabilistic model which we can apply to the task of image completion. Basically, we observe the top half of an image of a handwritten digit, and we’d like to predict what’s in the bottom half.