Hello,
I was wondering if it would be possible to store the feature vectors in a matrix during offline learning and use this matrix to process a batch analysis with the standard box in openvibe. I noticed that we need to wait for reading the signal we recorded during a learning session, which takes time even though we can fast forward it. Moreover generic stream file is quite heavy in memory in comparison to this matrix.
I suppose matlab boxes will be needed in case there is no box for that purpose.
Thank you in advance for your answer,
Nicolas.
Features concatenation as a matrix
Re: Features concatenation as a matrix
Dear Nicolas,
just save the extracted "feature vector" stream to a file using the generic stream writer box, then read it back (fast forwarding it) with the stream reader box in your other scenario
I hope this helps,
Best regards,
Yann
just save the extracted "feature vector" stream to a file using the generic stream writer box, then read it back (fast forwarding it) with the stream reader box in your other scenario
I hope this helps,
Best regards,
Yann