BEHIND THE SCENES OF TRAINING, MANAGING AND DEPLOYING MODELS

Abstract: “The model was working just fine two weeks ago, but now I can’t reproduce it!”

“Bob’s on vacation – how do I run his model?”

“Is my neural network useless or should I continue tweaking its parameters?”

Have you ever heard any of the above before? We had the same problems when running research and multiple commercial machine/deep learning projects. Based on our experience, we have distilled a number of best practices on team cooperation and model deployment, that can significantly improve your team’s performance. We will guide you through the process of building a robust data science pipeline by using a range of technologies (e.g. Git, Docker or Neptune – our in-house tool for managing machine learning experiments). Join our session and also share your best practices with us. Let’s do data science the right way!

Bio: Piotr is a data scientist, focusing on machine learning and data visualization. He holds PhD in quantum physics from ICFO, Barcelona. He lectures at Imperial College London and gave talks at Caltech and Bay Area D3.js User Group among other places. Piotr is the author a popular blog post series introducing to data science (“Data science intro for math/phys background”), word2vec (“king - man + woman is queen; but why?”) and neural networks (“Learning Deep Learning with Keras“, “Human log loss for image classification”). In free time he develops the Quantum Game with Photons and volunteers in teaching gifted high-school students.

Open Data Science Conference