Creating Custom Jupyter Kernels in TLJH

This tutorial shows non-admin users how to create their own custom Jupyter kernels with specific Python packages. No admin access required!

Method 1: Create a Kernel from requirements.txt

Use this method when you have a requirements.txt file with the packages you need.

Open a terminal in JupyterHub
Go to: File → New → Terminal
Create a virtual environment
```
python3 -m venv ~/my-custom-env
```
Replace my-custom-env with whatever name you prefer.
Activate the environment
```
source ~/my-custom-env/bin/activate
```
Install ipykernel (required!)
```
pip install ipykernel
```

Install your packages

pip install -r requirements.txt

Or install packages individually:

pip install pandas numpy matplotlib scikit-learn

Register the kernel with Jupyter

python -m ipykernel install --user --name=my-custom-env --display-name="My Custom Environment"

The --display-name is what you'll see in Jupyter's kernel menu.

Deactivate the environment
```
deactivate
```

Method 2: Register an Existing Virtual Environment

If you already have a virtual environment set up, you can register it as a kernel.

Activate your existing environment
```
source /path/to/your/venv/bin/activate
```
Install ipykernel if needed
```
pip install ipykernel
```

Register the kernel

python -m ipykernel install --user --name=my-project --display-name="My Project"

Deactivate
```
deactivate
```

Using Your Custom Kernel

Refresh your JupyterHub browser page
Create a new notebook or open an existing one
Go to: Kernel → Change Kernel
Select your custom kernel from the list

Managing Your Kernels

List all available kernels:

jupyter kernelspec list

Remove a kernel you no longer need:

jupyter kernelspec remove my-custom-env

View kernel location:

Your kernels are stored in: ~/.local/share/jupyter/kernels/

Important Notes:

The --user flag installs the kernel only for your account (no admin needed)
Each kernel is independent - you can have multiple kernels with different package sets
To update packages in a kernel, activate its virtual environment and use pip install --upgrade
Kernels survive JupyterHub restarts - you only need to create them once

Example: Creating a Data Science Kernel

Here's a complete example for a data science environment:

# Create and activate environment
python3 -m venv ~/ds-env
source ~/ds-env/bin/activate

# Install packages
pip install ipykernel
pip install pandas numpy scipy matplotlib seaborn scikit-learn jupyter

# Register kernel
python -m ipykernel install --user --name=ds-env --display-name="Data Science"

# Deactivate
deactivate