[Mne_analysis] Question regarding computation time of Temporal Generalization

Gabriel Della Bella gabyarg25 at gmail.com
Fri Oct 2 11:36:09 EDT 2020
Search archives:

        External Email - Use Caution        

Okay, that makes sense. I didn't know you could parallelize so easily!

Thanks Giulia!

El vie., 2 de oct. de 2020 a la(s) 12:29, Giulia Gennari (
giulia.gennari1991 at gmail.com) escribió:

>         External Email - Use Caution
>
> Hello Gabriel,
>
> I think it is normal: you are training 4000 classifiers!
> Any chance you can downsample a bit? 125Hz resolution might be good enough
> already.
> Also, it is very useful to use parallel jobs e.g. to try: n_jobs=*-1*
>
> Good luck!
>
> Giulia
>
> On Fri, Oct 2, 2020 at 5:17 PM Gabriel Della Bella <gabyarg25 at gmail.com>
> wrote:
>
>>         External Email - Use Caution
>>
>> Hi,
>> I'm interested in computing the Temporal generalization so I followed
>> this guide
>> (
>> https://mne.tools/0.17/auto_tutorials/plot_sensors_decoding.html#temporal-generalization
>> )
>> I'm using the exact same code as in the example. In my case, I have X
>> being a 30x256x4000 matrix and y is a 30x1 array (since I have 30 trials,
>> 256 channels and 4000 time samples)
>> The thing is that processing this code takes a long time, around 60 hs.
>> My pc specs are: Intel Core i7-7700 CPU @ 3.60 Hz x 8 and 32 GB of RAM.
>> Is this normal? Am I doing something wrong?
>>
>> Thank you!
>> Gabriel
>> _______________________________________________
>> 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/20201002/1ff22547/attachment-0001.html 


More information about the Mne_analysis mailing list