Exploring AutoML for Deep Learning with Ludwig

Learn how open-source Ludwig makes it easy to train state-of-the-art tabular and text classification models in minutes

Unstructured text data provides a wealth of information that can be used for a broad range of use cases from sentiment analysis to automated document topic recognition. But, developing effective deep learning models for text data can be difficult and expensive. Instead of writing hundreds of lines of code, you can use Ludwig – an open-source declarative machine learning framework – to create a model in just a few lines of a configuration file.  And you can use Ludwig AutoML to reduce the time and compute resources needed to develop a tuned model.
Ludwig’s simple and flexible data-driven configuration system makes it easy to rapidly build end-to-end deep learning pipelines in minutes. In experimental studies, the best models produced by Ludwig AutoML for tabular and text classification datasets were found to be competitive with published manually-tuned models.

In this webinar, we’ll deep-dive into Ludwig AutoML design, development, evaluation, and use. And we’ll provide links to example invocations for 36 different datasets, as well as include brief pointers on how you can enhance Ludwig AutoML capabilities.

Join us to learn:
  • The motivations behind adding AutoML to Ludwig, an open-source declarative ML toolkit created at Uber, and how Ludwig AutoML functions,
  • About Ludwig AutoML design, development, and evaluation, and
  • How to rapidly build tabular and text classification models with Ludwig AutoML

Watch the on-demand webinar

Anne Holler
Ludwig Contributor and previously Staff Engineer at Uber

Anne Holler is a leading contributor to Ludwig (www.ludwig.ai), the open-source declarative ML framework that makes it easy to build state-of-the-art ML pipelines with a simple and flexible data-driven configuration system. Previously, Anne was a Staff Engineer working at Uber on the Michelangelo machine learning platform and before that she worked at Velocloud, on the management stack for SD-WAN, and at VMware, on Distributed Resource Schedulers for server and storage.
Daliana Liu
Senior Data Scientist
Daliana is a Sr. Data Scientist at Predibase as well as the host of the popular Data Scientist Show podcast with over 200K followers on LinkedIn. Previously she was a Sr. ML Instructor and Data Scientist at AWS.