Pytorch, Scikit-learn, OpenCV, Flask
To learn more about computer vision, I built an American Sign Language letter classifier. I tried a number of models, from SVMs to different CNN architectures, and finally settled on a VGG-11. I also learned some data augmentation techniques.
After noticing a lack of transparency in company diversity statistics, my friends and I built a Chrome extension that makes this information more accessible, especially for those seeking a job. Our project won two awards at the TechTogether Boston 2019 hackathon.