Jupyterlab IDE

Getting Started with my Jupyter IDE

Your Jupyterlab IDE is a managed service that provides you access to preconfigured images of the latest versions of the Forest SDK. To get started with your IDE, just click the JupyterLab tab on the menu bar.

You will be greeted by your Jupyter Console:

After clicking Start My Server you will be given the option to select a pre-configured server to use:

Finally, you should be see the Jupyterlab interface, where you have access to a convenient IDE and terminal for iterative development:

Jupyterlab is a popular tool for iterative development. If you are new to Jupyter, checkout Jupyterlab user guide for an overview of what it hows and best ways to use it.

Create a virtual environment using virtualenv:

Your IDE comes with virtualenv installed allowing you to create virtual environments using the command below. It is recommended you use the --system-site-packages flag when creating your environment which allows the environment you are creating to inherit the base packages of the image you are using.

$ virtualenv --system-site-packages NAME_MY_ENVIRONMENT

You can activate your newly created environment using the following command:

$ source NAME_MY_ENVIRONMENT/bin/activate

Tip: For your default environment, you can add the statement below to your .bashrc file so that it is activated automatically every time you login.

$ source NAME_MY_ENVIRONMENT/bin/activate

Registering virtual environments as launcher shortcut:

In order to create a launcher shortcut, install ipykernel and use it to register your environment:

$ pip install ipykernel
$ python -m ipykernel install --user --name=NAME_MY_ENVIRONMENT

If this widget doesnt show up immediately, refresh your screen and open the launcher by clicking the + under the file tab.

Managing Launcher Kernels:

Use jupyter kernelspec list to see the list of existing kernels:

$ jupyter kernelspec list

Output: Available kernels: venv /home/jovyan/.local/share/jupyter/kernels/venv ir /opt/conda/share/jupyter/kernels/ir julia-1.5 /opt/conda/share/jupyter/kernels/julia-1.5 python3 /opt/conda/share/jupyter/kernels/python3

You can change the Kernel name on the launcher by following one of the paths above and editing the display_name in the `kernel.json` file

Removing a kernel from launcher

$ jupyter kernelspec remove old_kernel

Kernel specs to remove: old_kernel /home/jovyan/.local/share/jupyter/kernels/venv

Remove 1 kernel specs [y/N]: y

[RemoveKernelSpec] Removed /home/jovyan/.local/share/jupyter/kernels/venv