Building Experiment Tracking at Scale with Weights & Biases

The Data Science Discipline of the School of EECS is hosting the following guest seminar: 

Building Experiment Tracking at Scale with Weights & Biases

Speaker: Andrea Parker, Growth ML Engineer (Weights & Biases)
Host: Dr Yadan Luo

More Info: https://uq-ds-seminar.github.io/weights-biases/

Abstract: Building experiments doesn’t just end once the model is deployed. Teams need to monitor their models in production and use their findings to iterate further. Especially when dealing with tens to hundreds of models you need to monitor and automate at scale! How do you track your experiments? Have you ever found it hard to reproduce them and is sharing results with your colleagues and managers a headache? Join us to learn all about experiment tracking at scale with Pachyderm and Weights & Biases. W&B introduction and product walkthrough (talk will touch on: experiment tracking, W&B Tables, Sweeps, Artifacts, Dashboards/Reports, Integrations), followed by a Colab classification competition in Kaggle with swag for top submission. 

Speaker Bio: Andrea is a Machine Learning engineer who has worked extensively with large-scale graphs. She has a degree in Information Retrieval from the University of Michigan School of Information and has been working in the industry for over a decade. Andrea is passionate about data-driven problem solving and enjoys working on NLP and IR challenges across varied domains: health data, networks/linked data, and more.