W&B Events for the Week of October 25th
Publish your model insights with interactive plots for performance metrics, predictions, and hyperparameters. Made by Rajne Kumari using Weights & Biases
Oct 26th, 3:00pm PT - Chris Van Pelt at AWS Startup Show
Our co-founder, Chris Van Pelt was at the AWS Startups Show to discuss the steps and tools startup founders need to consider as they build out their AI/ML startup and lessons Chris learned along the way in his startup journey.
We dug into what MLOps means, what makes sense to track as you are building models, and why developer experience is so critical in achieving success in AI/ML.
Oct 27th, 10am PT - Chris Van Pelt at Fireside Chat: MLOps Ecosystem Panel
One of the most common hurdles with developing models is to design meaningful features that can operate at scale and be easily explained by anyone regardless of their level of domain expertise. Having access to a consistent set of fully contextualized features during different phases of the ML lifecycle is critical to building explainable ML/AI. During this talk, Chris Van Pelt demonstrated several best practices on how fellow data scientists can easily incorporate domain context into their feature engineering processes and make it easily discoverable by others. Following these guidelines will help teams transform their existing ad-hoc feature engineering methods into an explainable and repeatable process that can scale with the business.
Oct 30th, 2pm PT - Lukas Biewald, Reproducible Machine Learning at Scale
How can we support effective, reproducible, and explainable deep learning and coordination across practitioners? In this talk, Lukas Biewald shared best practices for conducting, debugging, and sharing deep learning experiments at scale. He talked through how some of the best tech companies in the world use the Weights & Biases platform for managing datasets, debugging models, versioning training/evaluation recipes, extracting insights, and storing all the crucial details needed to make their models reproducible and their research collaborative.