Hi

This github repository contains code changes to learn about writing a simple Jupyter Kernel called MyEcho Kernel. As the name suggests, the kernel will simply return whatever is passed from the Notebook UI. The kernel is a wrapper kernel over the existing ipykernel library. More details at writing swapper kernel.

Structure of the Project

1. echo_kernel directory contains the actual python code for starting the kernel python process.
2. kernel-spec directory contains the Spec file which will be installed in Jupyter Environment.
3. example directory contains a sample notebook which can be used to test the kernel.
   You are free to create a new notebook of your choice.
4. Makefile contains the handy commands which will to setup and test this project.
5. setup.py is a standard setup file for installing the project.
6. requirements.yml is a standard conda requirements file which contains the python dependencies that are needed.

Pre-requisite

1. git clone this repository:

git clone git@github.com:rahul26goyal/jupyter-echo-kernel.git
cd jupyter-echo-kernel

Setting up the Env using Make Commands.

This is option 1 where we start fresh from creating a conda environment and installing
all the libraries required to test this kernel on a Classic Jupyter Notebook UI.

1. Create a conda env: make env
2. Activate the env: make activate-env or conda activate learning_echo_kernel
3. Install the kernel library to environment. make dev-install
4. Install the kernel Spec to environment: make install-echo-kernel-spec
5. Verify the kernel spec is available : jupyter-kernelspec list
6. Install Classic Jupyter notebook: pip instal notebook==6.4.6
7. Start Jupyter Notebook serve: jupyter notebook --debug

install the kernel to separate notebook env

This is option 2.

• Lets assume you have a different conda env jupyter_notebook where notebook run
• active the conda env: conda acivate jupyter_notebook
• install this Echo kernel to the same env: make dev-install
• install the kernel-spec to the jupyter env:

make install-echo-kernel-spec
# verify the kernel is installed.
jupyter-kernelspec list

• start jupyter notebook : jupyter notebook --debug
• create a notebook by selecting "MyEchoKernel" or open the notebook present in the example/ folder.

Thank you reading this! If you have any feedback or suggestion, feel free to raise an issue.