server

ABIM

Hello

Is the process and data in server public?
how can I use server?( save the process and data in server)
Is the data in local repository save in our pc?

Regards
A

BalazsBarany

Hi @ABIM,

1. you decide where the process runs. If it is coming from a server, you'll see different start options in Studio: Locally, in the background (still in Studio), or on the server.

2. can you clarify what you mean with dynamically?
You can schedule the process on the server to run every hour or minute or twice a day, with a flexible schedule. Just select "Schedule on Server" in your Studio.
You can also expose the process on the server as a web service. In that case you can define macros in your process that will be parameters of the web service. You'll be able to call these web services from any application that has this functionality, or from scripting and programming languages.

Regards,
Balázs

BalazsBarany

Hi @ABIM,

I try to respond but I'm not entirely sure what you mean.

To use a server (nowadays called AI Hub), you need to deploy it. This is described in the documentation:
https://docs.rapidminer.com/latest/deployment/overview/
There are also Academy videos on the deployment: https://academy.rapidminer.com/learn/video/ai-hub-deployment-with-docker

When the server is running, you can connect Studio to it and store processes and data either into the main repository or into a project. In the repository there is a full access control infrastructure, you can specify who can access what. With projects you can control who can use the project, but inside the repository there is no finer access control by design. So you (or the administrator of your AI Hub) can decide what is public and what isn't.

If you use a local repository in Studio, that is by default a folder structure on your PC. You can put it on a network drive to work together with colleagues, though. For this, you just create a new repository and select the network path. Others will simply add this path in their Studios as a new repository to collaborate.

I hope this is what you meant with your questions. Feel free to clarify if I got you wrong.

Regards,
Balázs

ABIM

@BalazsBarany

Thank you for your reply. 
I have a server and I want to write a webservice from my process and call it to my program. Now questions are these:
1.How can I connect my process to my server with a web service? (with using my server)
2.Is the temporary repository a kind of cloud for RM team server? and if we have any process there, who can access to that process and data? (how can I manage it?)

3.If I delete any process or data from temporary repository, is it delete from your server?

Regards
A

BalazsBarany

Hi @ABIM,

web services are defined in the server interface under Processes/Web Services. There you select an existing process from the repository and define inputs (parameters) that will be passed to the process as macros.
https://docs.rapidminer.com/latest/hub/web-services/create-web-service.html

The output of the process can be converted to different formats: XML, JSON etc. This might be relevant if you want to use the process from RapidMiner Studio or a different application. 

The AI Hub repository is persistent. If you put something into it, it will stay there until you delete it. I don't know of a temporary repository implementation. Which problem are you trying to solve?  

There is something you might want to use, but is meant for web apps, not independent services. There are operators called Publish to App and Recall from App. These have temporary contents (in memory), identified by a name, valid in a user session. They can be used for communicating between different subprocesses, or in a web app. You could try to check if they solve your problem.

Regards,
Balázs

ABIM

@BalazsBarany

1.I am new to rapidminer server. I run a process in studio and now I want to know If I run the process in server, is it run in my own server or it will run in RM server?( because it is for a project and I want to use my own server to save the data and process there)

2. After running the process in studio, how can I use it in my server dynamically?

Thank you very much 
A

ABIM

@BalazsBarany

Hello

Thank you for your answers, they were useful. For dynamically your answer is correct and that was my goal.
Now I have 3 repositories in my RM. I know the local is in my pc but I don't have any information about other ones. And just I find a post in community about local and server. Now please tell me what are these repositories in screen?( what is legacy?)


Regards
A

BalazsBarany

Hi!

Samples is a local read-only repository with samples, maintained by RapidMiner.
Community Samples is online and also read only.
Keras Samples comes from the Deep Learning extension.

As far as I know, there was one change in repositories recently: The legacy ones are in $HOME/.RapidMiner/repositories (so, essentially hidden), the new ones are in $HOME/Documents/RapidMiner, so you can easily find them in your file management application.

Local Repository is created automatically upon installation.
The Temporary Repository is used for stuff like AutoModel. 
New Local Repository was created by you.

There is also a different type of repositories, Projects. These are managed with the Git version control software.

I use different kinds of repositories without problems, I don't think you'll have problems with these.

Regards,
Balázs

ABIM

Thank you very much!