by

Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Jan 31 2019 22:33
    codecov[bot] commented #5895
  • Jan 31 2019 22:32
    larsoner synchronize #5895
  • Jan 31 2019 21:16
    codecov[bot] commented #5895
  • Jan 31 2019 21:16
    larsoner synchronize #5895
  • Jan 31 2019 20:23
    agramfort commented #5768
  • Jan 31 2019 18:31
    larsoner opened #5895
  • Jan 31 2019 18:31
    larsoner milestoned #5895
  • Jan 31 2019 17:58
    codecov[bot] commented #5822
  • Jan 31 2019 17:16
    GuillaumeFavelier commented #5822
  • Jan 31 2019 17:13
    codecov[bot] commented #5822
  • Jan 31 2019 17:13
    GuillaumeFavelier synchronize #5822
  • Jan 31 2019 17:01
    jshanna100 commented #5807
  • Jan 31 2019 16:11
    larsoner commented #5807
  • Jan 31 2019 15:36
    jshanna100 commented #5807
  • Jan 31 2019 15:19
    codecov[bot] commented #5807
  • Jan 31 2019 15:07
    GuillaumeFavelier commented #5822
  • Jan 31 2019 14:18
    codecov[bot] commented #5768
  • Jan 31 2019 14:01
    codecov[bot] commented #5807
  • Jan 31 2019 14:01
    jshanna100 synchronize #5807
  • Jan 31 2019 13:45
    GuillaumeFavelier synchronize #5822
demitau
@demitau
@larsoner ok, I've done it mne-tools/mne-python#7856 Its my first time opening an issue on GitHub so I don't really know what is the right way to format it, sorry for that.
nadiamoazen
@nadiamoazen

HI ,I I followed the MNE Installation instruction and
I test it:

mne sys_info
Platform: Windows-10-10.0.18362-SP0
Python: 3.8.3 (default, May 19 2020, 06:50:17) [MSC v.1916 64 bit (AMD64)]
Executable: c:\users\nadia\anaconda3\envs\mne\python.exe
CPU: Intel64 Family 6 Model 142 Stepping 11, GenuineIntel: 8 cores
Memory: 15.8 GB
mne: 0.20.5
numpy: 1.18.1 {blas=mkl_rt, lapack=mkl_rt}
scipy: 1.4.1
matplotlib: 3.1.3 {backend=Qt5Agg}
sklearn: 0.22.1
numba: 0.49.1
nibabel: 3.1.0
cupy: Not found
pandas: 1.0.3
dipy: 1.1.1
mayavi: 4.7.2.dev0 {qt_api=pyqt5, PyQt5=5.9.2}
pyvista: 0.24.2
vtk: 9.0.0

Since My platform is window10, I didn't install freeSurfer.
However, after running code, it still gives an error "No module named 'mne'"
could you help how can I solve this error?

Daniel McCloy
@drammock
@nadiamoazen the first thing I would check is that you activated the conda environment before running your code
nadiamoazen
@nadiamoazen
@drammock
In Anaconda prompt,I commanded" conda activate mne",
then "I conda install spyder"
I also add C:\Users\nadia\Anaconda3\envs\mne\python.exe
to python interpreter.
Christian O'Reilly
@christian-oreilly
In what coordinate system (e.g., RAS) are the coordinates in montage.dig?
Somehow this information seems hard to find in the doc (or I was unlucky in search)
Daniel McCloy
@drammock
@nadiamoazen in Spyder, Tools > Preferences > Python interpreter, and Tools > Preferences > Run are the next places I would look to see if / why the script might not be running in the right environment. It sounds like you already checked the first of these... any Windows experts care to chime in? I'm running out of ideas.
Eric Larson
@larsoner
they can be in native space if lpa/nasion/rpa are defined in the same space
they get converted to head coords when you do inst.set_montage(montage)
head coords meaning Neuromag RAS head coords
Christian O'Reilly
@christian-oreilly
Ok, thanks for the info @larsoner
Stephan Heunis
@jsheunis
Hi everyone. Sorry if this is the wrong forum for this question. I would like to invite a leading developer/advocate of MNE Python to give a short demo-flavoured talk in the virtual Open Science Room of the annual OHBM meeting happening later this month. It would need to be someone who is already registered for or planning to attend OHBM, and who has time to record a 7min talk this week. The Open Science Room schedule has been in flux during the last days because of a few cancellations, hence the new opening and limited time for recording. Please let me know if you have suggestions for someone to contact. I'm happy to answer questions to shed more light on this. Thanks!
Alexandre Gramfort
@agramfort
@jsheunis I am registered for OHBM but I already have another talk to do this for OHBM + other stuff. I would love someone else to do this video but if nobody volunteers I can try
Mpriour42
@Mpriour42

Hi everyone ! I preprocessed my EEG raw data using MNE functions and created a dictionnary for all files containing information about related bad epochs and bad channels. I save it as a .pkl file using a pickle function. This "preprocessed.pkl" file was the basis for all my further analyses and it was working well before. However, now when I try to open it using the pickle.load :
import mne
import pyconscious as pc
import pickle
import numpy as np

filename = 'preprocessed.pkl'
loadfile = open(filename,'rb')
fileinfo = pickle.load(loadfile)
loadfile.close()

I got an error message "ModuleNotFoundError: No module named 'mne._digitization'". I don't understand where this comes from and when I preprocessed my data again all like before it is working, but I would like to be able to open my former preprocessing file for my analyses to go on the same basis.

Thank you very much for your attention to my bug. I hope you can help me figure this out !

Stephan Heunis
@jsheunis
@agramfort great, thanks for the interest. Do you know who else I could reach out to?
Alexandre Gramfort
@agramfort
@jsheunis maybe @dengemann would be happy to help?
Denis A. Engemann
@dengemann
I’m happy to help @jsheunis
Stephan Heunis
@jsheunis
@dengemann awesome! I'll dm you
Denis A. Engemann
@dengemann
@jsheunis @agramfort I don’t have calm on my side the next weeks to do pre-recordings. Perhaps @jasmainak could be interested ?
Denis A. Engemann
@dengemann
cc @wmvanvliet
Stephan Heunis
@jsheunis
@dengemann @agramfort It doesn't make a huge difference, but today the decision was made to extend the recording slots to 11 June. If this changes things for you, please let me know. Otherwise I'm happy to follow up with @wmvanvliet as well.
Marijn van Vliet
@wmvanvliet
@jsheunis I could do it. Basically a 7 minute infomercial for MNE-Python?
@dengemann @agramfort and everyone: what would you like to see highlighted?
For me, that would be: MNE-Python is more than a piece of software, it's a community.
Marijn van Vliet
@wmvanvliet
The new Pyvista brain viz of course
Stephan Heunis
@jsheunis
@wmvanvliet yes that's a good description :)
I'll follow up via email
Stephan Heunis
@jsheunis
Just in case anyone else will help or be involved: the idea is to have a 7min presentation on MNE python. It's in the virtual Open Science Room, for which the sessions will be broadcast publicly. We will prerecord talks. You can use the OHBM-contracted option for which you have to book a slot asap for doing the recording, with slots available until 11 June. You can also record it yourself, then send us the recording by the same date. We ask that the speakers be available for short Q&A sessions on the day that the recording will be aired. This will be on 25 June. The recording will be shown three times, once per time-zone hub (Asia+Pacific; Europe+Middle East+Africa; Americas). You can also suggest a representative to stand in for Q&A if you are not available for some of the three airings (which would be understandable given time-zone challenges)
Stefan Appelhoff
@sappelhoff
hooray for @wmvanvliet :-)
agpr141
@agpr141
Hello. I am able to use raw.plot() with a rawEDF file but not when I load in a rawFIF using mne.read_raw_fif. With the FIF, python backend crashes. Is there a workaround round this?
Alexandre Gramfort
@agramfort
thanks heaps @wmvanvliet !
@agpr141 what version of mne are you using? what OS? please report mne.sys_info()
agpr141
@agpr141

Hi @agramfort
Platform: Darwin-19.4.0-x86_64-i386-64bit
Python: 3.7.6 (v3.7.6:43364a7ae0, Dec 18 2019, 14:18:50) [Clang 6.0 (clang-600.0.57)]
Executable: /Users/Amber/PycharmProjects/analysis/bin/python
CPU: i386: 4 cores
Memory: Unavailable (requires "psutil" package)
mne: 0.20.5
numpy: 1.17.3 {blas=openblas, lapack=openblas}
scipy: 1.4.1
matplotlib: 3.2.0 {backend=module://backend_interagg}
sklearn: 0.23.1
numba: 0.49.1
nibabel: Not found
cupy: Not found
pandas: 1.0.4
dipy: Not found
mayavi: Not found
pyvista: Not found
vtk: Not found

MacOS Catalina 0.15.4

Thanks!

Marijn van Vliet
@wmvanvliet
Other things that come to mind:
Continuous HPI tracking
Marijn van Vliet
@wmvanvliet
Overhauled tutorials
@agpr141 is there an error message somewhere that could give us a clue?
Marijn van Vliet
@wmvanvliet
I'm looking for photos of our hackathons for use as a backdrop for the talk about our wonderful community
Does anyone have some?
Eric Larson
@larsoner
@wmvanvliet Matti took photos, I'll ping him
Marijn van Vliet
@wmvanvliet
:thumbsup:
agpr141
@agpr141
@wmvanvliet No error message comes up, the Python icon just jumps up and down on the dock before exiting and closing the python console on PyCharm
Marijn van Vliet
@wmvanvliet
I'm trying to think what the difference is between plotting the EDF file and the FIFF file. After loading, MNE-Python generally cannot tell the difference between them. Is the FIFF file much larger (longer, more channels) than the EDF file?
Britta Westner
@britta-wstnr
@wmvanvliet there is this photo on twitter from Paris 2019: https://twitter.com/agramfort/status/1121026403871408130 -- tweet by @agramfort
Marijn van Vliet
@wmvanvliet
Nice one, I'll definately use it
Eric Larson
@larsoner
@agpr141 I first would try to eliminate PyCharm as being the problem by running the plotting code directly in python from the command line, there you should see PyQt5 as the default backend (hopefully), and if things work, then it suggests that the backend_interagg is the problem
Adam Li
@adam2392

Hi, I had a quick performance question that I couldn't figure out from looking at the documentation. Was hoping someone just had a ball-park answer.

So when raw is loaded, if you don't use preload, then things can be done very fast regardless of how large the dataset is (e.g picking, thanks Eric for that sweet PR). So if you are say running a parallel computation across contiguous windows of the raw dataset, would passing around the reference to raw be preferable?

e.g. when you say do raw.get_data(start=window_start, stop=window_end) will this be the same speed regardless of how big the whole raw dataset is?

I wanted to see what the tradeoff of passing around the raw reference, versus loading in the entire dataset and passing around the specific contiguous chunks.

Lmk if I can elaborate?

Eric Larson
@larsoner

when you say do raw.get_data(start=window_start, stop=window_end) will this be the same speed regardless of how big the whole raw dataset is?

Depends on what you mean. In general we try to make it so that e.g., reading 1 sec of data from anywhere in the data file takes the same amount of time. But then if you want to read the first second and the last second of data and you are using actual spinning disks (not SSD), it will be slower because of the time required to fid.seek to a different part of the file (I think? Also OS caching, etc. aside). If you expect to have to read the same segments of data over and over again, you're probably better off preloading everything before you do that, as raw.get_data will just be a memcopy, more or less (extremely fast). If you know you are going to need all or most of the data, doing raw.load_data() then doing raw.get_data(start, stop) over and over again will be faster than doing on-demand reads, though OS file caching in memory might not make this true, depending on your system.

TL;DR: A lot of variables at play for what is slow vs fast depending on what you actually want to do, if performance is critical, try it each way on some limited set of data and see the extent to which it matters

Jean-Rémi KING
@kingjr

Hello!

I'm facing difficulties rendering 3d figures on a headless machine using Xvfb :1 -screen 0 1280x1024x24 -auth localhost & export DISPLAY=:1

Running

from mayavi import mlab
mlab.options.offscreen = True
mlab.test_contour3d()
mlab.savefig('example.png')

works well, but stc.save_image('test.png') makes a black image.

Any idea on how to debug ?