My research interests are in Deep Learning, and application of Machine Learning in Healthcare and Public Health. I also educate high school students about STEM.
Develop Machine Learning apps using Dash and Plotly. Projects completed include interactive visualization of high-dimensional datasets, object detection enhancement app, and real-time model training viewer.
Applied Machine Learning Research at the Surveillance Lab, an Epidemiology lab within the McGill University Faculty of Medicine. The paper was presented at the 2018 AAAI Joint Workshop on Public Health Intelligence, which you can find here.
You can find below a sample of my projects. Click here to see the complete list.
For every Deep Learning models, keeping track of accuracy and loss is an essential part of the training process, since they indicate how good your models are. This app is a real-time visualization app that monitors core metrics of your Tensorflow graphs during the training, so that you can quickly detect anomalies within your model.Demo App | Project Repo
This app provides useful visualizations about what's happening inside a complex video in real-time. The data is generated using MobileNet v1 in Tensorflow, trained on the COCO dataset. The video is displayed using the community-maintained video component.Demo App | Project Repo
This app wraps Pillow, a powerful image processing library in Python, and abstract all the operations through an easy to use GUI. All the computation is done backend through Dash, and image transfer optimized through session-based Redis caching and S3 storage.Demo App | Project Repo