3. Installation

Note

For questions related to installing or using RMG please post an issue on the RMG-Py GitHub repository. You can also search for your problem on the issues page to see if there are already solutions in development. Alternatively, you can email us at rmg_dev@mit.edu

3.1. Recommended Install: Docker

RMG is primarily distributed using Docker, a software package for delivering applications.

1. Download and install Docker.

2. Open a terminal, powershell, or command prompt and run docker pull reactionmechanismgenerator/rmg:3.2.0.

   This step may take some time as the image is downloaded.

3. Run docker run --name rmgcontainer -v "C:\Users\rmguser\myrmgfiles:/rmg/RMG-Py/myrmgfiles" -it reactionmechanismgenerator/rmg:3.2.0

   This command will make the folder C:\Users\rmguser\myrmgfiles on your computer accessible from inside the container to easily edit and transfer input and output files. Change the path to match your individual computer. If the folder does not exist when the command is run, it will be created.

   If you want to use jupyter notebook inside the docker container, run docker run --name rmgcontainer -v "C:\Users\rmguser\myrmgfiles:/rmg/RMG-Py/myrmgfiles" -it -p 8888:8888 reactionmechanismgenerator/rmg:3.2.0 instead. And you can start the jupyter notebook by running jupyter notebook --ip 0.0.0.0 --no-browser --allow-root inside the container. Then you can access the jupyter notebook from your browser by going to http://localhost:8888. You may need to copy and paste the token from the terminal into the browser to access the notebook.

You are now operating inside an Ubuntu operating system (a container called “rmgcontainer”) with a working installation of RMG-Py. To leave this container run exit, and to reconnect run docker start rmgcontainer --attach --interactive.

For users unfamiliar with bash or Linux, we recommend looking at online Linux tutorials particularly Linux vs. Windows, Terminal vs File Manager, and Must Know Linux/Unix Commands.

3.2. Alternative Install: Binary Installation Using Anaconda

If you are accustomed to using the Anaconda package manager or cannot tolerate the storage overhead of Docker, installation from conda is also available. This is recommended for users who want to use RMG out of the box and are not interested in changing the RMG code or making many additions to RMG’s thermodynamic and kinetics databases. If this does interest you, please see the Developer Install below.

• 3.2.1. Binary Installation Using Anaconda for Unix-Based Systems: Linux and Mac OSX

3.3. Developer Install: Installation from Source

RMG-Py can now be built by source using the Anaconda Python Platform to assist in installing all necessary dependencies. This is recommended for a developer who may be altering the RMG source code or someone who expects to manipulate the databases extensively. You will also be able to access the latest source code updates and patches through Github.

• 3.3.1. Installation by Source Using Anaconda Environment for Unix-based Systems: Linux and Mac OSX
• 3.3.2. Updating the RMG-Py Source Code

3.4. Archive of Unsupported Installation Methods

Below are old installation techniques that are no longer supported, including instructions for installation without using Anaconda and the old installation instructions for Windows. These instructions are no longer maintained, and are not recommended for use.

• 3.4.1. Linux Installation
• 3.4.2. MacOS X Installation
• 3.4.3. Binary Installation Using Anaconda for Windows
• 3.4.4. Installation by Source Using Anaconda Environment for Windows
• 3.4.5. Setting up Windows Environment Variables for RMG
• 3.4.6. Setting up a Linux Virtual Machine from Windows
• 3.4.7. Installing RMG in the Linux Subsystem on Windows 10

3.5. Dependencies

Please visit the page below for detailed information on all of RMG’s dependencies and their license restrictions

• 3.5.1. Dependencies