Yida Wang

The description makes sense to me. I couldn't remember all the details of how fslmaths works. If it is as described above, I am fine with the proposed fix with...

I don't quite remember the numbers I got before. H]On how many TRs do you compute the correlation? Can you split the 3 s into correlation, normalization and classification? Thanks!...

Setting the `random` argument to 1 or 2 shuffles the data not the labels. http://brainiak.org/docs/brainiak.fcma.html#brainiak.fcma.preprocessing.RandomType What is your dataset? What is the classification results you got from `random=0`?

@mihaic sounds good, I can do it later today.

Thanks for investigating this, @amennen Within the same set of test samples, can the distance value indicate the confidence?

It depends on the latest advance of the area. We will cover some on-going research topics at the end.

It looks like that your system doesn't have MPI packages. I would recommend you check out the FCMA component of [BrainIAK](https://github.com/IntelPNI/brainiak), which has better support and should be easier to...

You may check if the MPI library is correctly linked while compiling FCMA. It seems to me that the system doesn't find corresponding libraries even if you have installed somewhere.

Aldo, If you would like to leverage HPC to do some correlation computation quickly, I would recommend you to refer to FCMA@BrainIAK again. We have example code for high-performance correlation...