A key step in the data science workflow is rapid model development in order to create, test, and identify the best models to put into production. However, large gaps exist in this workflow, and the data science tool set is rapidly changing to fill those gaps. Large teams and enterprises are quickly moving from using individual siloed notebooks like Zeppelin and Jupyter to wanting to share and reuse models, code, and results. Challenges also exist in deploying models into production and model serving using tools like Kubeflow and TensorFlow. Moon Soo Lee and Louis Huard explore real-world examples of how companies are solving these problems, and how you can use these best practices in your own workflow.
What you'll learn
- Learn how companies are solving the problem of the gaps in the data science workflow
This session is from the 2019 O'Reilly Artificial Intelligence Conference in San Jose, CA.
- Title: The holy grail of data science: Rapid model development and deployment (sponsored by Zepl)
- Release date: February 2020
- Publisher(s): O'Reilly Media, Inc.
- ISBN: 0636920369936
You might also like
Head First Design Patterns, 2nd Edition
You know you don’t want to reinvent the wheel, so you look to design patterns—the lessons …
Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd Edition
Through a series of recent breakthroughs, deep learning has boosted the entire field of machine learning. …
Analytical Skills for AI and Data Science
While several market-leading companies have successfully transformed their business models by following data- and AI-driven paths, …
Software Engineering at Google
Today, software engineers need to know not only how to program effectively but also how to …