MI-Classification Alternative algorithms

Making & changing box plugins and external apps
Post Reply
mhadji05
Posts: 53
Joined: Tue Mar 16, 2021 9:37 am

MI-Classification Alternative algorithms

Post by mhadji05 »

Hello all.
Happy New Year.

I m trying to improve my MI (Left-right hand imagination) system with better classification algorithms.

Classifier Trainer box support only LDA, SVM, MLP.
I want to try XG Boost, RF, grid search for example.

Is there any chance to add more algorithms?
If no. How can I do it by my self?

Thomas
Posts: 210
Joined: Wed Mar 04, 2020 3:38 pm

Re: MI-Classification Alternative algorithms

Post by Thomas »

Hi mhadji05,

Happy New Year to you too.

It is definitely possible to add more algorithms.
OpenViBE is open source and we definitely welcome contributions from anyone.

Before modifying the code, I invite you to look at the compilation instructions.

Once this is done, you will find tutorials on how to modify OpenViBE on our documentation page.
The section you are interested in is under Modifying OpenViBE -> Box Plugins.
The classification plugin boxes files are in the following path: sdk/plugins/processing/classification/src/.

If you need any further information, please do not hesitate.

Looking forward to see some new algorithms integrated to OpenViBE. :D

Cheers,
Thomas

mhadji05
Posts: 53
Joined: Tue Mar 16, 2021 9:37 am

Re: MI-Classification Alternative algorithms

Post by mhadji05 »

Hi Thomas

Actually, I don't think I can deploy an openvibe box. I wish I could.

My question was if the openvibe development team has any plans to develop new algorithms? Please let me know.

Alternatively, is there a repository where I can find the algorithms I mentioned above in python (Random Forest and XG boost, linked with grid search classifier hyperparameter tuning)?

Basically I would like to leverage the scikit learn in openvibe for MI (Left-Right hand imagination) which provides these algorithms (Random Forest and XG boost) but I don't know how.

Can you guide me how can I achieve this?

Post Reply