[Mne_analysis] Memory error reading raw files

Karin Westin karin.westin85 at gmail.com
Sun Dec 16 15:03:40 EST 2018
Search archives:

        External Email - Use Caution        

Ok thanks!


Den sön 16 dec. 2018 kl 19:18 skrev Eric Larson <larson.eric.d at gmail.com>:

>         External Email - Use Caution
>
> Or you can use memmapping by passing a filename string to `preload`, which
> will be used for temporary storage.
>
> Eric
>
>
> On Sun, Dec 16, 2018, 11:37 Alexandre Gramfort <
> alexandre.gramfort at inria.fr wrote:
>
>>         External Email - Use Caution
>>
>> hi Karin,
>>
>> loading the file in memory will double the size as we do computations
>> in double precision.
>>
>> I would see if you can use preload=False on read. The only thing that
>> can be problematic
>> is filtering which requires preload=True. In this case, I would
>> process data by chunks. You can
>> use the raw.crop method for this.
>>
>> HTH
>> Alex
>>
>>
>>
>> On Sun, Dec 16, 2018 at 5:22 PM Karin Westin <karin.westin85 at gmail.com>
>> wrote:
>> >
>> >         External Email - Use Caution
>> >
>> > Hi!
>> > I'd like to preload raw-fif files (~40 min recording, sample rate 5000
>> Hz, 10.2 GB) but it seems I'm running out of memory (RAM 16GB, 2x8 GB). Do
>> I simply need more memory or is there any other way to tweak the loading of
>> the raw files?
>> >
>> > Best,
>> > Karin
>> > _______________________________________________
>> > Mne_analysis mailing list
>> > Mne_analysis at nmr.mgh.harvard.edu
>> > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis
>>
>> _______________________________________________
>> Mne_analysis mailing list
>> Mne_analysis at nmr.mgh.harvard.edu
>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis
>>
> _______________________________________________
> Mne_analysis mailing list
> Mne_analysis at nmr.mgh.harvard.edu
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20181216/71851ae2/attachment.html 


More information about the Mne_analysis mailing list