Jupyter Newsletter 10• December 21, 2016
nbdime 0.1 and nbconvert 5.0 released
nbdime 0.1
We’ve just published the first release of nbdime, which makes it easier to work with notebooks in version control system such as Git. More specifically, nbdime provides tools for:
- Viewing diffs of notebooks in the terminal;
- Viewing rich diffs in a browser;
- Automatically resolving merge conflicts; and
- Three-way merging of notebooks in a browser.
Check it out, let us know how it goes, and help out if you can! This work is led by Martin Alnæs and Vidar Fauske at Simula Research Laboratory, and supported by the OpenDreamKit project, funded by the EU H2020 program.
nbconvert 5.0.0
We’ve released nbconvert 5.0.0 which includes a few major changes and plenty of bug fixes. One of the key changes was switching the default engine for generating PDFs from pdfLaTeX to XeLaTeX. XeLaTeX allows better unicode and font handling, microtypography, and opens the way for future work to support a wider range of language scripts (among other things).
There is also a new AsciiDoc Exporter, a preprocessor for sanitizing html (for use, for example, in Wikimedia), and a more general pandoc interface. This interface allows using pandocfilters to directly manipulate the pandoc’s internal JSON structure for documents, which allows extensions to inspect and alter notebook content in robust ways not possible before. There have been loads of other bugfixes and improvements over the past year, especially in the ways nbconvert interacts with pandoc and LaTeX. You can read about all this in greater detail in the changelog.
Upgrade today with:
pip install -U nbconvertLearn, Create and Collaborate with IBM Data Science Experience
Wondering how to get started with data science? The IBM Data Science Experience (‘DSX’) is a useful tool for getting ramped up with data science. It offers ready-to-use open-source tools such as the Jupyter Notebook and RStudio (in a single location!) along with helpful content in the form of starter notebooks, tutorials and sample data. DSX is an interactive, cloud-based environment where users can share access to data and common workspaces (‘Projects’) in order to collaborate and learn together.
Curious about DSX? Check out how a user analyzed Game of Thrones using DSX: https://apsportal.ibm.com/blog/game-of-thrones-with-dsx/
See what DSX has to offer in terms of features and a free trial option at: http://datascience.ibm.com/
Community Based Jupyter Notebook Tutorial
Karlijn Willems, a data science journalist has created an easy-to-follow tutorial on DataCamp for beginners that explains how to install and run the Jupyter Notebook Application and how it can be used as an interactive environment for data science. The tutorial, entitled “Jupyter Notebook Tutorial: The Definitive Guide,” includes tips and best practices to use Jupyter Notebooks efficiently on your own or in data science teams, and examples of Jupyter Notebooks that will inspire new users.
This tutorial joins DataCamp’s ever growing content for learning data science in Python and R. While in New York City last Fall, Jupyter contributor Brian Granger met with DataCamp’s Hugo Bowne-Anderson to talk about their data science learning platform. Recent online courses at DataCamp cover DataFrames, Bokeh, Anaconda, and Statistical Thinking in Python.
Featured Community Members
Charnpreet Singh is Software Engineer for Jupyter at Cal Poly. When he’s not working on Jupyter, he has a passion for developing iOS applications. He co-founded and is currently the Vice President of Cal Poly’s Mobile App Development Club. At Jupyter he contributes to frontend development on JupyterLab and the Jupyter website. His latest project is to create a custom Sphinx theme for the project to use in all of its documentation.
You can find him on GitHub as @charnpreetsingh.
Sylvain Corlay is a quant researcher specializing in stochastic analysis. He is an adjunct faculty member at the Courant Institute and Columbia University where he teaches numerical methods for finance.
As an open source developer, Sylvain mostly contributes to Project Jupyter in the area of interactive widgets and lower level components such as traitlets. Besides Jupyter, Sylvain contributes to a number of other open-source projects for scientific computing and data visualization, such as spyder, bqplot, pythreejs and ipyleaflet.
Sylvain founded QuantStack in September 2016. Prior to founding QuantStack, Sylvain was a quant researcher at Bloomberg LP in New York City.
twitter: https://twitter.com/SylvainCorlay
github: https://github.com/SylvainCorlay
Recent events
JupyterDay Paris, December 6, 2016. Our fifth and final JupyterDay for 2016 was held in Paris, France. A few core developers including Thomas Kluyver, Sylvain Corlay, Min Ragan-Kelley, and Afshin Darian were on hand to share their knowledge with the audience. Special thanks go out to The LoOPS network, DevLog, and OpenDreamKit for organizing the event.
Learning Jupyter, December 13, 2016. PyLadies and Bloomberg recently hosted an evening of knowledge sharing, installs, and data science. Jamie Whitacre, Technical Project Manager at Project Jupyter presented a preview of JupyterLab and other Jupyter subprojects in the development pipeline during the event held in San Francisco, California. She also covered how to install the R kernel, and how run Python and R in the same notebook.
Upcoming events
AnacondaCON, February 7–9, 2017. The inaugural Anaconda conference will take place in Austin, Texas in a couple months. To learn more visit https://anacondacon17.io
