I have two variables as a result of an ICA (Independent Component analysis) decomposition (decomposed using joint diagonalization algorithm which uses frobenius norm formulation) named A_ffdiag and W_ffdiag of shapes A_ffdiag : (116, 30) ,W_ffdiag : (30, 116) (where 30 is the number of components and 116 is the number od channels).

Now, I believe that some operations need to me made on these two variables to convert them to mixing/unmixing matrix and further use them as ICA instance in MNE(python library). Can somebody help me in solving how can I convert these two variables into mixing/unmixing matrix?

  • $\begingroup$ Could you clarify a bit for us? $\endgroup$
    – Gillespie
    May 11, 2022 at 12:44
  • $\begingroup$ How are you calculating A_ffdiag and W_ffdiag? Are they outputs from some software package run on your data? What package? Have you checked the documentation of the function that gave you these outputs? My guess would be they are the mixing and unmixing matrices, respectively. But it's hard to tell because your question is vague. $\endgroup$
    – Gillespie
    May 11, 2022 at 12:51


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Browse other questions tagged or ask your own question.