[Eeglablist] Double precision

[Makoto Miyakoshi]

Dear Jumana,

> I am not using the GUI so do I need to specific double precision every
time I reload the data?

Yes, you still need to change the EEGLAB option.

> Also, as I batch process 600 subjects, I save the data after nearly every
major processing step.

That's a big number of datasets!

> Unfortunately I did not realise the  file options were ticket to single
precision. Therefore I filtered the data and referenced to a single
electrode, deleted bad segments of data and then saved, loaded the data and
then ran ICA.
> I realise that pop runica converts to double precision, but would I have
lost information in previous process of steps by having the 'single'
default ticked?
> Do I therefore need to redo my analysis? And if so, how should I concert
to double in code? And do I do it after every save, or after every
computation?

You don't need to worry about it. Single precision becomes a problem when
you deal with matrix operations such as pinv() etc. For standard processes,
using single precision is ok; infomax was automatically performed as double
precision as you mentioned, which is a requirement!

Makoto



On Sun, Dec 4, 2016 at 3:23 AM, Ahmad, Jumana <jumana.ahmad at kcl.ac.uk>
wrote:

> Dear all,
>
> I understand the rationale for using double precision, but am confused as
> to when to convert the data. I am not using the GUI so do I need to
> specific double precision every time I reload the data?
>
> Also, as I batch process 600 subjects, I save the data after nearly every
> major processing step.
>
> Unfortunately I did not realise the  file options were ticket to single
> precision. Therefore I filtered the data and referenced to a single
> electrode, deleted bad segments of data and then saved, loaded the data and
> then ran ICA.
>
> I realise that pop runica converts to double precision, but would I have
> lost information in previous process of steps by having the 'single'
> default ticked?
>
> Do I therefore need to redo my analysis? And if so, how should I concert
> to double in code? And do I do it after every save, or after every
> computation?
>
> Best wishes,
> Jumana
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.
> ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Makoto Miyakoshi
Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20161206/1af50a61/attachment.html>