Chris Winsor
Machine Learning, Industrial IoT, Machine Vision, Robotics
I am a software developer focused on Industrial IoT, machine vision, robotics and machine learning. My toolset includes PyTorch, Torch Geometric, transfer learning, pretrained models, Colab, Django and Precise Automation robotics.
I like to keep on top of the latest ML techniques and “kick the tires” through proof-of-concept experiments. I’m always looking for projects like the ones below so free to reach out!
Chris (cwinsor@gmail.com)
Graph Neural Network using PyG (Pytorch Geometric) In this project we model topic evolution on social media using a Twitter dataset of Covid-19 tweets. We use PyTorch Geometric, Sequence Encoder, HeteroConv and a BERT language model.
Whole-image Classification using Kiras/Tensorflow: This is a self-driving car that (in realtime) classifies images from a front-mounted camera. It is an end-to-end CNN deployed to Raspberry Pi.
StarChaser: This is a web-app to demonstrate Machine Learning integrated into an every-day Website/app. It uses Django, Postgres and a little JavaScript. Presentation, Game
Markov Logic Network is a new and exciting technique for modeling systems that have structure but are inherently probabilistic.
Kaggle PLAsTiCC: A survey of approaches for classification of an aperiodic timeseries dataset from Kaggle.
Tri-Training: Use unlabeled data for supervised learning! The paper is from an Amazon researcher in Cambridge MA and was used for wake-word detection but the approach can be applied anywhere.
The Bootstrap: A principled technique to establish statistical significance to test results. Combined with T-tests this is important to have in the tool belt.
Industrial Internet-of-Things (IIOT)
Probabilistic Graphical Models