# Converting ICA decomposition with 2 variables A\_ffdiag and W\_ffdiag into MNE mixing/unmixing matrix

**URL:** <https://mne.discourse.group/t/converting-ica-decomposition-with-2-variables-a-ffdiag-and-w-ffdiag-into-mne-mixing-unmixing-matrix/4819>\
**Category:** Support & Discussions\
**Created:** [May 2, 2022, 11:58am UTC](https://mne.discourse.group/t/converting-ica-decomposition-with-2-variables-a-ffdiag-and-w-ffdiag-into-mne-mixing-unmixing-matrix/4819 "2022-05-02T11:58:30Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Anand](https://avatars.discourse-cdn.com/v4/letter/a/ecd19e/32.png) [@Anand](https://mne.discourse.group/u/Anand)\
**Post date:** [May 2, 2022, 11:58am UTC](https://mne.discourse.group/t/converting-ica-decomposition-with-2-variables-a-ffdiag-and-w-ffdiag-into-mne-mixing-unmixing-matrix/4819/1 "2022-05-02T11:58:30Z")

</div>

Hi,  
I am looking to convert ICA decomposition (using joint diagonization algorithm) which have 2 variables `A_ffdiag` and `W_ffdiag` of shapes A\_ffdiag : (116, 30) ,W\_ffdiag : (30, 116) into the MNE mixing/unmixing matrix in the ICA instance. Can somebody help me in solving how can I convert these two variables into mixing/unmixing matrix?  
Thanks