Efficient AutoML with Ludwig, Ray, and Nodeless Kubernetes

Efficient AutoML with Ludwig, Ray, and Nodeless Kubernetes

Explore how open-source platforms Ludwig and Ray democratize Deep Learning by simplifying the process of training, scaling, deploying, and serving models. Dive into Ludwig's recent AutoML extensions for tabular and text classification datasets, leveraging Ray Tune for efficient hyperparameter search. Learn about the heuristics employed by Ludwig AutoML to produce effective models for validation datasets. Discover the cost-saving and operational benefits of running Ludwig AutoML on cloud Kubernetes clusters with Nodeless K8s, which dynamically allocates and removes GPU resources as needed, compared to running directly on EC2 instances.