- Plotly.py 3.0.0 has been released and it comes with tons of new features and changes
- Major updates – Jupyter widget support, an imperative method for making changes on-the-fly, remarkably detailed attribute descriptions
- Other changes include specific error messages, a blazing fast execution, exporting static images, among other things
Python is undeniably the most popular language being used by data scientists globally. It has eaten up the ground on R in the last couple of years and is rapidly ascending to scale new heights. One of the best things about R has been it’s ggplot2 package, which has always been my go-to tool for visualization (even that has now been introduced in Python).
Python’s data visualization library catalogue is also diverse, though nothing had been as good as ggplot2. Until now. Here comes the latest release in the popular plotly.py library, 3.0.0, and it’s a big one! There are tons of new features and changes that will make working with this library even more interactive and enjoyable.
You can download plotly using the below commands:
pip install plotly
In case you already have plotly installed in your machine, you can upgrade it by typing the below command:
pip install plotly --upgrade
Let’s look at what all is new in this release.
- Jupyter Widget Support: A new Jupyter widget class, plotly.graph_objs.FigureWidget, is included in this release. This is tailor made specifically for Jupyter Notebooks and JupyterLab environments. You can even zoom into specific regions in the plot!
- Imperative Method: This release contains a set of methods that allow you to manipulate and explore the plots in Jupyter. This is called the imperative method, which basically means that you can edit your plots on-the-fly inside your Jupyter notebook
- Detailed Attribute Descriptions: It’s now easier than ever to understand the attributes on your plots without having to leave your notebook. Each attribute has a rich description attached with it
- Other Changes: Errors are now displayed with more specific messages, you can now define transition animations with ContextManager, generate and export static images of plotly plots
Plotting thousands on points in your notebook used to take a bit of time. This has also been vastly improved in plotly 3.0.0 and you will notice how quickly charts are popping up on your machine.
You can read more about these changes, and see them in action, via Plotly’s Twitter account.
Our take on this
Plotly’s release came at approximately the same time that ggplot2 3.0.0 was launched. It’s a great time to be a data visualization user! These changes in plotly will greatly enhance my experience with visualizations (I am already a heavy visualization user). I’m really looking forward to using the Jupyter widget support and see how that functions in my notebooks.
Another thing worth exploring is the ‘Imperative Method’ which gives us far more flexibility in drawing the plots (and not just how they look). You can even use plotly with R if you with Shiny apps. If you do use this latest release, be sure to let me know your experience in the comments section below!
Subscribe to AVBytes here to get regular data science, machine learning and AI updates in your inbox!