Global AI and Data Science

Global AI & Data Science

Train, tune and distribute models with generative AI and machine learning capabilities

 View Only

Jupyter Track @ Strata SF : M Pacer, Chris Holdgraf

By William Roberts posted Thu May 02, 2019 03:35 PM

  
This year at Strata San Francisco, IBM sponsored a series of talks in a track designed for people using Jupyter Notebooks. Chris Holdgraf (UC Berkeley Institute for Data Science) and M Pacer (Netflix) were two of the many excellent speakers in that track, and they both shared valuable presentations. 


For the uninitiated, Project Jupyter is the open source project focused on providing a web-based interface that can host comments, markup, as well as execute Python. Chris Holdgraf elegantly would explain it is also "a community of people and an ecosystem of open tools and standards for interactive computing." It's a powerful tool for those who need to do exploratory data analysis, create plots, and then share those results. Besides the graphical user interface, which makes it easy to share visualizations, the notebooks themselves are easily shareable - all you need is a copy of that file. Both Chris and M offer their compelling use cases. 


M Pacer's talk offered a view of how one would use notebooks for semantically rich conversations. As a Jupyter notebook holds state (is a record), it is the basis of communicating shared representations between people (conversation), but importantly M also teaches the stack in which it is easy for a person to communicate to/from a notebook. Check out M Pacer's talk if you're ready to learn this superpower. 


Chris Holdgraf shared how UC Berkeley has built its ecosystem to teach its undergraduate students in data science using Jupyter book. Chris and the Institute for Data Science believe in the radical idea that some data science should be taught to everyone, but UC Berkeley has many students so how do you build a scalable stack for that many tenants? Jupyter does much to provide an interactive environment based on an open stack. Check out Chris Holdgraf's talk if you're ready to see projects that enable that solution, and what it would look like for any enterprise struggling to connect many users with content and computation.


If you're curious about Project Jupyter and how IBM is involved, as well as other open source projects, check out IBM's CODAIT (Center for Open-Source Data & AI Technologies). In the meantime, for thought leadership in Data Science, join IBM's Data Science community today!

#GlobalAIandDataScience
#GlobalDataScience
0 comments
21 views

Permalink

https://community.ibm.com/community/user/blogs/william-roberts/2019/05/02/jupyter-track-strata-sf-m-pacer-chris-holdgraf