There have been some major advancements to the RapidMiner platform since this article was originally published. We’re on a mission to make machine learning more accessible to anyone. For more details, check out our latest release.
Where did my Processes Go?!
You’ve built your workflow processes, linked up your databases, and put your models in production. Things are running so smoothly and then you get a notification that RapidMiner has released a newer and cooler version of RapidMiner Studio and RapidMiner AI Hub (formerly RapidMiner Server).
You freeze.
Should I upgrade? What’s going to happen to all my processes and data? Locally and on the RapidMiner AI Hub? Will everything stop working?
You’re stressing!
Stop stressing – it’s all good!
The reality is that nothing happens to your processes and data. They all remain where they are on your disk. Upgrades won’t affect them at all.
Where’s my Process?
Once you install RapidMiner Studio for the first time, it will automatically create a local repository on your machine. A local repository is where you store your processes and any data you import onto your machine. For Window’s users, this location is under //.RapidMiner/repositories. Mac and Linux users will find this under their /home/.RapidMiner folders.
Note: The installation directory of RapidMiner Studio is always different from where the user stores his/her processes and data.
When working with RapidMiner Studio you can create as many new repositories (e.g., for new projects) and they will all be stored under the default /.RapidMiner/repositories folder location. You can even specify a customer folder (i.e. /Documents/my-processes/) to save your processes and data in.
Upgrading RapidMiner Studio
When you upgrade RapidMiner Studio, your repositories are not affected because they’re tied to your user directory. The upgrade only affects where the RapidMiner Studio program is installed.
Note: Don’t create a custom directory for your repository in the same location where the RapidMiner Studio program is installed. In that situation, an upgrade could wipe out your repository.
What about Compatibility?
With every new version of RapidMiner Studio, things get better. More machine learning algorithms, better memory core management, etc. Sometimes individual operators are enhanced or modified. When that happens you’ll see a little icon at the bottom of the Operator’s parameter window. In the example below we show the Series Extension Windowing operator.
It was recently updated for version 7.3 from version 5.2. If your process relies on using an older version, you can just click on “Change compatibility” to the previous version.
The little triangle with an exclamation point is just a visual cue to make sure you really want to use the older version of the operator.
Note: Sometimes there are compatibility issue with 3rd party non-RapidMiner supported extensions. If you have a lot of them installed, check with the 3rd-party developer to ensure their extension is compatible with the latest version of RapidMiner first!
Moving your User Profile
Sometimes you need to move your user directory to another computer or machine. You can just migrate that profile to the new machine, install RapidMiner Studio again, and then manually edit the location of the repository in RapidMiner Studio.
The way to do that is to right click on the repository in the repository pane, select Configure Repository, and update the path.
Upgrading RapidMiner AI Hub (formerly RapidMiner Server)
All your processes and data stored on RapidMiner AI Hub is saved in a database. When you installed RapidMiner AI Hub, it requires a database to connect to. That’s where RapidMiner AI Hub does all its Cron jobs, queuing, dashboarding, and where it stores your processes & data.
When you upgrade RapidMiner AI Hub, you normally download just the EAR file, shutdown RapidMiner AI Hub, do a database backup, and then drop in the new EAR file in the /standalone directory. Then you restart it and all your processes and data will populate from there.
See? No need to worry about upgrading to the latest and greatest. We’ve got you covered.
There have been some major advancements to the RapidMiner platform since this article was originally published. We’re on a mission to make machine learning more accessible to anyone. For more details, check out our latest release.
