Intro to Deep Learning with Keras (Carpentries)

About this resource

The Intro to Deep Learning with Keras workshop from the Carpentries will walk you through introductory deep learning concepts as well as how to build a neural networks in Keras. Keras is high-level wrapper framework (uses PyTorch or Tensorflow in the backend) which allows you to train and evaluate neural networks in just a few lines of code. It may take slightly longer to train a Keras model (compared to PyTorch and Tensorflow), but the difference in performance is often negligible for those that only need to train a few models. The ability to quickly build and test models is the primary selling point of Keras.

Prerequisites

Learners are expected to have the following knowledge:

• Basic Python programming skills and familiarity with the Pandas package. If you need a refresher, these Introductory Python lesson materials are available for independent study.
• Basic knowledge on machine learning, including the following concepts: Data cleaning, train & test split, type of problems (regression, classification), overfitting & underfitting, metrics (accuracy, recall, etc.).The Intro to Machine Learning with Sklearn lesson materials are a good option for those that need a refresher on machine learning fundamentals.

Estimated time to complete

This workshop takes approximately 15 hours to complete.

Register to take this workshop in Madison!

The Carpentries is a global organization of researchers who volunteer their time and effort to create workshops that teach software engineering and data analysis skills to other researchers. UW-Madison has its own local Carpentries community which is actively engaged in developing new ML/AI workshops. To be notified of upcoming workshops offered by the Carpentries, make sure to subscribe to the Data Science @ UW Newsletter. The Intro Deep Learning workshop is typically taught in May each year.

Alternatively, work through the materials independently!

All Carpentries lessons are published as open source educational materials. You are welcome and encouraged to visit the lesson materials to work through them on your own. If you are involved with a research lab at UW-Madison campus, you may attend Coding Meetup (Tue/Thur, 2:30-4:30pm) to get help working through the materials.

Questions?

If you any lingering questions about this resource, please feel free to post to the Nexus Q&A on GitHub. We will improve materials on this website as additional questions come in.

See also

• Workshop: Intro to Deep Learning with PyTorch: Explore PyTorch as an alternative deep learning framework.
• Book: Understanding Deep Learning - Simon J.D. Prince: This free textbook is a good modern overview of deep learning, and provides colab notebooks to explore deep learning concepts and implementations. The book uses PyTorch as its framework of choice. You may find additional details in this book that the workshop only briefly touches on.