Webinar: "Boosting Models and Time Series Forecasting Models with W&B"
Publish your model insights with interactive plots for performance metrics, predictions, and hyperparameters. Made by Andrea Pessl using Weights & Biases
Though deep learning is all the rage, we know different, more traditional modeling techniques are still the bedrock for many ML-powered services today. And W&B helps with those too!
In this webinar, we explored work shared by Paul Edwards and Stephen Denton from ScotiaBank at Nvidia GTC21 on how to build a credit scorecard using boosting models such as XGBoost.
See how it can improve performance over handcrafted scorecards while still remaining interpretable for regulators.
We also explored some of the functionality of LinkedIn Engineering's new GreyKite library for time series forecasting that was recently released, including trend visualization & model fine-tuning.
💡And if you can't make it to our session, don't worry: the recording will be shared with all sign-ups.