Mar 15–18, 2020

Using Keras to classify text with LSTMs and other ML techniques

lukas biewald (Weights & Biases)
1:30pm5:00pm Monday, March 16, 2020
Location: 210 D/H

Who is this presentation for?

  • Machine learning engineers




You’ll go hands-on to build a sentiment classifier on text data starting with scikit-learn and Naive Bayes and continuously improving the model.

Lukas Biewald explains Keras and applies word embeddings and convolutional neural networks (CNNs). Along the way, you’ll build a variety of LSTMs and GRUs, and you’ll be given a JupyterHub instance setup with a cloud GPU to follow along.

Prerequisite knowledge

  • Familiarity with Python
  • Experience with ML (not necessarily on text)

Materials or downloads needed in advance

  • A WiFi-enabled laptop with the Chrome web browser installed
  • Everything will be hosted on a JupyterHub in a web browser to avoid installation problems. If you really want to configure things locally, install requirements.txt or follow the README.

What you'll learn

  • Learn text handing with scikit-learn
  • Understand how to build LSTMs and GRUs
  • Get an introduction to Keras
Photo of lukas biewald

lukas biewald

Weights & Biases

Lukas Biewald is the founder and CEO of Weights & Biases, his second major contribution to advances in the machine learning field. In 2009, Lukas founded Figure Eight, formally CrowdFlower. Figure Eight was acquired by Appen in 2019. Lukas has dedicated his career to optimize ML workflows and teach ML practitioners, making machine learning more accessible to all.

Leave a Comment or Question

Help us make this conference the best it can be for you. Have questions you'd like this speaker to address? Suggestions for issues that deserve extra attention? Feedback that you'd like to share with the speaker and other attendees?

Join the conversation here (requires login)

Contact us

For conference registration information and customer service

For more information on community discounts and trade opportunities with O’Reilly conferences

Become a sponsor

For information on exhibiting or sponsoring a conference

For media/analyst press inquires