The Yhat Blog


machine learning, data science, engineering


Rodeo 0.4: Spark, themes, resizeable panes

by Greg |


Well it's been over a month since we launched Rodeo. It's been really exciting seeing how people have been using it and the sorts of features they've been asking for. Lucky for them, we've been listening! Today we're announcing the second major release of Rodeo.

We've added some great features for things like Spark, autocompletion, and enhancing the customization of the Rodeo UI. Hope you like them.

Install / Upgrade

First off, if you haven't already, go ahead and install/upgrade to the latest version of Rodeo. You can do this using pip:

bash $ pip install --upgrade rodeo

Spark Integration

PySpark is all the rage these days, so it was of little surprise when this was one of the first PR's we got on GitHub. Big thanks to Ben Cook from Hudl for submitting this.

The Spark integration allows you to connect Rodeo to a Spark cluster. Configure your Spark environment variables, (SPARK_HOME, SPARK_OPTS, SPARK_LOG), then fire up PySpark and then pass the --pyspark flag to the rodeo command.

Using the sparkcontext, sc, module from Rodeo.

Resizeable panes

One big feature request that we got from the initial launch was support for resizeable panes. We've added in the ability to resize the 4 panes that appear in the UI. We've also made the app more responsive, so those of you with smaller screens will be able to enjoy Rodeo in it's full glory.

Personalized Configs

Another big gripe about Rodeo was that the font size was too big/small, or that the syntax highlighting wasn't quite right. To combat this we've added a few configurable UI features: color scheme, font-size, and keyboard shortcuts (which includes VIM and Emacs modes).

Autocomplete

Another feature that's made a lot of progress is the autocomplete. Previously, Rodeo was using Jedi which is a really cool project, but it felt a little awkward since we were using the IPython kernel for everything else.

We've switched over entirely to the IPython autocompleter and with it fixed some UI defects regarding the autocompletion. What's really cool about this is that by switching entirely over to IPython, we're now one step closer to being able to support languages other than Python.

Luckily for us, the work for non-Python languages has largely been done by the great folks at the IPython project--we're just following in their footsteps. In the next release, we're going to roll out beta support for a new programing language! (exact language is still TBD)

Looking Forward

There's still a lot of work to be done, but the core of Rodeo is fairly stable. In the coming releases, get ready for an improved UI, more configuration options, and better support for server-side deployments.



Our Products


Rodeo: a native Python editor built for doing data science on your desktop.

Download it now!

ScienceOps: deploy predictive models in production applications without IT.

Learn More

Yhat (pronounced Y-hat) provides data science solutions that let data scientists deploy and integrate predictive models into applications without IT or custom coding.