JavaScript Error: IPython is not defined [Solved]

Introduction

When working with an application within the browser, there might be a need for some JavaScript environment configurations. Applications such as JupyterLab are one of such applications that need JavaScript support, especially when running interactive code or plots.

In situations where the right environment is not set up in JupyterLab leading to errors. One such error is the Javascript Error: IPython is not defined error message.

In this article, we will discuss how to deal with the JavaScript Error message involving IPython.

 

Use matplotlib inline to solve JavaScript Error in IPython

In a situation where you are running a plot (using matplotlib) in a JupyterLab environment and encounter a Javascript Error: IPython is not defined error message, you can configure your matplotlib to use the inline magic command which allows you to draw only static plots within your JupyterLab.

To make use of the inline magic command, you can add the statement below before your plot code.

%matplotlib inline

However, this means that you won’t be able to have an interactive plot within your JupyterLab. If you need an interactive plot, the ipympl package is helpful.

 

Use ipympl to solve JavaScript Error in IPython

Unlike the inline magic command, you can make use of a matplotlib package named ipympl which provides Jupyter integration and allows you to work with interactive plots within your matplotlib.

To have that, you need to install ipympl using either pip or conda depending on whether you are using a native python environment or an anaconda environment respectively.

To use pip, you can use the command below

pip install ipympl

However, if you use conda, you can use the command below

conda install -c conda-forge ipympl

Also, if you are using JupyterLab 2, you might need to follow the command below

conda install -c conda-forge nodejs
jupyter labextension install @jupyter-widgets/jupyterlab-manager jupyter-matplotlib

After installing the ipympl package, you can restart your JupyterLab and make use of the widget magic command to allow for interactive matplotlib within your code.

%matplotlib widget

 

Summary

To run interactive plots using matplotlib, you will need an extension - ipympl - to provide for such which relies on NodeJS to work, or you can render such a plot as a static plot using the inline magic command.

 

References

GitHub - matplotlib/ipympl: Matplotlib Jupyter Integration
https://matplotlib.org/ipympl/