Please note that this page is under development, but feel free to check out some of my interesting side-projects below!
N.B. - Select academic projects can be found under Education > Courses
Putting Machine Learning to use in industry requires a lot more than just training a good model. Follow along with my end-to-end project to deploy an application that predicts customer churn for a telecom company. We build a Flask API that pulls our model from S3 and feeds prediction to a React frontend!
In this mini-project, we look at how we can automatically download YouTube audio from a query using FFmpeg and transcribe it using AssemblyAI's API. We go on to use these transcriptions to perform sentiment analysis for product reviews using a pretrained BERT model!
CAEs extend the concept of autoencoders to utilize convolutional layers. Click here to follow along with a Jupyter notebook exploring the use of CAEs on the MNIST handwritten digit dataset - we conclude by looking at their potential use for denoising images prior to classification!
GANs are composed of two neural networks which compete with the goal of generating data similar to a set of training data. Click here to learn more about GANS and how they can be used to generate handwritten digits based off the MNIST dataset!