---
description: Run, develop, and share data science projects using JupyterLab and Docker
keywords: getting started, jupyter, notebook, python, jupyterlab, data science
title: Data science with JupyterLab
toc_max: 2
summary: |
  Use Docker to run Jupyter notebooks.
tags: [data-science]
languages: [python]
aliases:
  - /guides/use-case/jupyter/
params:
  time: 20 minutes
---

Docker and JupyterLab are two powerful tools that can enhance your data science
workflow. In this guide, you will learn how to use them together to create and
run reproducible data science environments. This guide is based on
[Supercharging AI/ML Development with JupyterLab and
Docker](https://www.docker.com/blog/supercharging-ai-ml-development-with-jupyterlab-and-docker/).

In this guide, you'll learn how to:

- Run a personal Jupyter Server with JupyterLab on your local machine
- Customize your JupyterLab environment
- Share your JupyterLab notebook and environment with other data scientists

## What is JupyterLab?

[JupyterLab](https://jupyterlab.readthedocs.io/en/stable/) is an open source application built around the concept of a computational notebook document. It enables sharing and executing code, data processing, visualization, and offers a range of interactive features for creating graphs.

## Why use Docker and JupyterLab together?

By combining Docker and JupyterLab, you can benefit from the advantages of both tools, such as:

- Containerization ensures a consistent JupyterLab environment across all
  deployments, eliminating compatibility issues.
- Containerized JupyterLab simplifies sharing and collaboration by removing the
  need for manual environment setup.
- Containers offer scalability for JupyterLab, supporting workload distribution
  and efficient resource management with platforms like Kubernetes.

## Prerequisites

To follow along with this guide, you must install the latest version of [Docker Desktop](/get-started/get-docker.md).

## Run and access a JupyterLab container

In a terminal, run the following command to run your JupyterLab container.

```console
$ docker run --rm -p 8889:8888 quay.io/jupyter/base-notebook start-notebook.py --NotebookApp.token='my-token'
```

The following are the notable parts of the command:

- `-p 8889:8888`: Maps port 8889 from the host to port 8888 on the container.
- `start-notebook.py --NotebookApp.token='my-token'`: Sets an access token
  rather than using a random token.

For more details, see the [Jupyter Server Options](https://jupyter-docker-stacks.readthedocs.io/en/latest/using/common.html#jupyter-server-options) and the [docker run CLI reference](/reference/cli/docker/container/run/).

If this is the first time you are running the image, Docker will download and
run it. The amount of time it takes to download the image will vary depending on
your network connection.

After the image downloads and runs, you can access the container. To access the
container, in a web browser navigate to
[localhost:8889/lab?token=my-token](http://localhost:8889/lab?token=my-token).

To stop the container, in the terminal press `ctrl`+`c`.

To access an existing notebook on your system, you can use a
[bind mount](/storage/bind-mounts/). Open a terminal and
change directory to where your existing notebook is. Then,
run the following command based on your operating system.

{{< tabs >}}
{{< tab name="Mac / Linux" >}}

```console
$ docker run --rm -p 8889:8888 -v "$(pwd):/home/jovyan/work" quay.io/jupyter/base-notebook start-notebook.py --NotebookApp.token='my-token'
```

{{< /tab >}}
{{< tab name="Windows (Command Prompt)" >}}

```console
$ docker run --rm -p 8889:8888 -v "%cd%":/home/jovyan/work quay.io/jupyter/base-notebook start-notebook.py --NotebookApp.token='my-token'
```

{{< /tab >}}
{{< tab name="Windows (PowerShell)" >}}

```console
$ docker run --rm -p 8889:8888 -v "$(pwd):/home/jovyan/work" quay.io/jupyter/base-notebook start-notebook.py --NotebookApp.token='my-token'
```

{{< /tab >}}
{{< tab name="Windows (Git Bash)" >}}