Skip to content

Latest commit

 

History

History
 
 

matlab

README.md

Use MATLAB Integration for Jupyter in a Docker Container

The Dockerfile in this repository builds a Docker® image based on jupyter/base-notebook which contains the jupyter-matlab-proxy package and installs MATLAB® using the MATLAB Package Manager (mpm).

The resulting Jupyter environment enables you to start MATLAB in a web browser tab from a Jupyter® notebook.

If you want to install the MATLAB Integration for Jupyter without using Docker, use the installation instructions in the repository MATLAB Integration for Jupyter instead.

Requirements

  • Linux® Operating System
  • Docker
    • Version newer than 20.10.9 if building for ubuntu 22.04
  • MATLAB version newer than R2020b

Build Instructions

Get Sources

# Clone this repository to your machine
git clone https://github.com/mathworks-ref-arch/matlab-integration-for-jupyter.git

# Navigate to the downloaded folder
cd matlab-integration-for-jupyter/matlab

Build & Run Docker Image

# Build container with a name and tag of your choice.
docker build -t matlab-notebook .

Start the container, and forward the default Jupyter web-app port (8888) to the host machine:

docker run -it --rm -p 8888:8888 matlab-notebook

Customize the Image

The Dockerfile supports the following Docker build-time variables:

Argument Name Default value Effect
MATLAB_PRODUCT_LIST MATLAB Specify the list of products to install using product names separated by spaces. Replace spaces within names with underscores. For example: MATLAB Simulink Deep_Learning_Toolbox Parallel_Computing_Toolbox
See MPM Documentation for more information.
MATLAB_RELEASE r2023a The MATLAB release you want to install. MUST be newer than r2020b and specified in lower-case
LICENSE_SERVER unset The port and hostname of the machine that is running the Network License Manager, using the port@hostname syntax. For Example: 27000@MyServerName.
Click Using the Network License Manager to learn more.

Customize Products to Install

The Dockerfile defaults to installing MATLAB with no additional toolboxes or products into the /opt/matlab folder.

To customize the build, use the Docker build-time variable MATLAB_PRODUCT_LIST to specify the list of products you want to install into the Image.

docker build --build-arg MATLAB_PRODUCT_LIST="MATLAB Deep_Learning_Toolbox" -t matlab-notebook .

Build an Image for a Different Release of MATLAB

# Builds an image with MATLAB R2020b
docker build --build-arg MATLAB_RELEASE=r2020b -t matlab-notebook .

Build an Image with License Server Information

Including the license server information with the docker build command avoids having to pass it when running the container.

# Build container with the License Server
docker build  --build-arg LICENSE_SERVER=27000@MyServerName -t matlab-notebook .

Access the Jupyter Notebook by following one of the URLs displayed in the output of the docker run command. For instructions about how to use the integration, see MATLAB Integration for Jupyter.

Advanced

  • Installing MATLAB into the Docker image can make the image very large (greater than 10GB) depending on the MATLAB toolboxes installed. To make the image smaller, you can give the Docker container access to a MATLAB installation using a volume or bind mount. For more details see Provide MATLAB as a Volume or Bind Mount.

  • Use Dockerfile.byoi to build an image based on an existing Docker image with MATLAB in it. See README-byoi.md

Feedback

We encourage you to try this repository with your environment and provide feedback – the technical team is monitoring this repository. If you encounter a technical issue or have an enhancement request, send an email to jupyter-support@mathworks.com.