If you study or work with Data Science in Python, Jupyter Notebook is an essential tool. In this tutorial, we will set up a Jupyter Notebook server using Docker, providing a ready-to-use environment without complex manual configurations.

Requirements

To reproduce this environment, you must have Docker Engine installed on your computer.
To install Docker, follow the instructions for your operating system here: Install Docker.

Step 1: Create the Folder Structure

Open a terminal and run:

mkdir -p ~/dev/docker/jupyter-server/work
cd ~/dev/docker/jupyter-server

Step 2: Create the Dockerfile

We will create a Dockerfile based on the minimal-notebook Jupyter image. Other options are available here.

Create a Dockerfile with the following content:

FROM jupyter/minimal-notebook

COPY ./requirements.txt /tmp/requirements.txt
RUN pip install -r /tmp/requirements.txt

Step 3: Create the requirements.txt

List the required Python libraries for your projects in requirements.txt:

numpy==2.2.4
pandas==2.2.3
openpyxl==3.1.5
matplotlib==3.10.1

Step 4: Create the docker-compose.yml

The docker-compose.yml file simplifies service configuration. The work directory will be used to store notebooks.

services:
  jupyter:
    container_name: jupyter-server
    build: 
      context: .
    user: root
    volumes:
      - ./work:/home/jovyan/work
    ports:
      - 8888:8888
    command: "start-notebook.py --ServerApp.root_dir=/home/jovyan/work"

Step 5: Start the Server

Run the container:

docker compose up -d

Loaded URL on the browser

Step 6: Access the Server

To get the access URL, run:

docker logs jupyter-server 2>&1 | grep "127." | tail -1

Copy the displayed URL and paste it into your browser, example:

http://127.0.0.1:8888/lab?token=e8d99ec57344fc92060a598a5ae6656c5f162aa4511dc0da

Loaded URL on the browser

Tip: Using Jupyter in VSCode

If you use VSCode, you can connect to the Jupyter server to run notebooks directly in the IDE. Just enter the Jupyter URL when running notebook cells for the first time.

Configurar servidor Jupyter no VSCode

Configurar servidor Jupyter no VSCode

Step 7: Stop the Server

To stop the Jupyter Notebook container:

docker compose down

Now, you have a Jupyter Notebook environment set up with Docker, ready for Data Science projects! 🎯