<p><span style="padding: 3px 10px; border-radius: 5px; color: #ffffff; font-weight: bold; display: inline-block; background-color: #ff0000;"> External Email - Use Caution </span></p><p></p><div>What you're showing is the output of the plot_opm_data.py example script. What I suggested was to re-install MNE-Python and its dependencies and examine the output of *the reinstallation process*, not the output of the example script. Did you do that? What (if any) errors did you see during reinstallation? How did you do the reinstallation (pip, conda, something else)? Did you search for problems related to matplotlib and ft2font? What does mne.sys_info() tell you?<br></div><div><br></div><div class="protonmail_signature_block"><div class="protonmail_signature_block-user"><div>-- dan<br></div><div>Daniel McCloy<br></div><div>https://dan.mccloy.info<br></div><div>Research Scientist<br></div><div>Institute for Learning and Brain Sciences<br></div><div>University of Washington<br></div></div><div class="protonmail_signature_block-proton protonmail_signature_block-empty"><br></div></div><div><br></div><div>‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐<br></div><div> On Saturday, April 4, 2020 7:21 PM, Saeed Zahran <saeedzahran@hotmail.com> wrote:<br></div><div> <br></div><blockquote class="protonmail_quote" type="cite"><p><span style="background-color:rgb(255, 0, 0)"><span style="color:rgb(255, 255, 255)"><b> External Email - Use Caution </b></span></span><br></p><p><br></p><div style="font-family: Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);"><span>Thank you Dan for your reply, I did exactly what you recommended, still have the below error:</span><br></div><div style="font-family: Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);"><span></span><br></div><div style="font-family: Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);"><div><span>C:\Users\zahransa\anaconda3\python.exe C:/Users/zahransa/PycharmProjects/opm/plot_opm_data.py<br> </span> </div><div>Opening raw data file C:\Users\zahransa\mne_data\MNE-OPM-data\MEG\OPM\OPM_SEF_raw.fif...<br></div><div>Isotrak not found<br></div><div> Range : 0 ... 700999 = 0.000 ... 700.999 secs<br></div><div>Ready.<br></div><div>Current compensation grade : 0<br></div><div>Reading 0 ... 700999 = 0.000 ... 700.999 secs...<br></div><div>Filtering raw data in 1 contiguous segment<br></div><div>Setting up low-pass filter at 90 Hz<br></div><div><br></div><div>FIR filter parameters<br></div><div>---------------------<br></div><div>Designing a one-pass, zero-phase, non-causal lowpass filter:<br></div><div>- Windowed time-domain design (firwin) method<br></div><div>- Hamming window with 0.0194 passband ripple and 53 dB stopband attenuation<br></div><div>- Upper passband edge: 90.00 Hz<br></div><div>- Upper transition bandwidth: 10.00 Hz (-6 dB cutoff frequency: 95.00 Hz)<br></div><div>- Filter length: 331 samples (0.331 sec)<br></div><div><br></div><div>Setting up band-stop filter from 49 - 51 Hz<br></div><div><br></div><div>FIR filter parameters<br></div><div>---------------------<br></div><div>Designing a one-pass, zero-phase, non-causal bandstop filter:<br></div><div>- Windowed time-domain design (firwin) method<br></div><div>- Hamming window with 0.0194 passband ripple and 53 dB stopband attenuation<br></div><div>- Lower passband edge: 49.00<br></div><div>- Lower transition bandwidth: 0.50 Hz (-6 dB cutoff frequency: 48.75 Hz)<br></div><div>- Upper passband edge: 51.00 Hz<br></div><div>- Upper transition bandwidth: 0.50 Hz (-6 dB cutoff frequency: 51.25 Hz)<br></div><div>- Filter length: 6601 samples (6.601 sec)<br></div><div><br></div><div>Trigger channel has a non-zero initial value of 256 (consider using initial_event=True to detect this event)<br></div><div>201 events found<br></div><div>Event IDs: [257]<br></div><div>Traceback (most recent call last):<br></div><div> File "C:/Users/zahransa/PycharmProjects/opm/plot_opm_data.py", line 51, in <module><br></div><div> evoked.plot()<br></div><div> File "C:\Users\zahransa\anaconda3\lib\site-packages\mne\evoked.py", line 293, in plot<br></div><div> time_unit=time_unit, sphere=sphere, verbose=verbose)<br></div><div> File "<decorator-gen-134>", line 21, in plot_evoked<br></div><div> File "C:\Users\zahransa\anaconda3\lib\site-packages\mne\viz\evoked.py", line 733, in plot_evoked<br></div><div> time_unit=time_unit, sphere=sphere)<br></div><div> File "C:\Users\zahransa\anaconda3\lib\site-packages\mne\viz\evoked.py", line 208, in _plot_evoked<br></div><div> import matplotlib.pyplot as plt<br></div><div> File "C:\Users\zahransa\anaconda3\lib\site-packages\matplotlib\pyplot.py", line 2282, in <module><br></div><div> switch_backend(rcParams["backend"])<br></div><div> File "C:\Users\zahransa\anaconda3\lib\site-packages\matplotlib\pyplot.py", line 221, in switch_backend<br></div><div> backend_mod = importlib.import_module(backend_name)<br></div><div> File "C:\Users\zahransa\anaconda3\lib\importlib\__init__.py", line 127, in import_module<br></div><div> return _bootstrap._gcd_import(name[level:], package, level)<br></div><div> File "C:\Users\zahransa\anaconda3\lib\site-packages\matplotlib\backends\backend_tkagg.py", line 2, in <module><br></div><div> from .backend_agg import FigureCanvasAgg<br></div><div> File "C:\Users\zahransa\anaconda3\lib\site-packages\matplotlib\backends\backend_agg.py", line 50, in <module><br></div><div> from PIL import Image<br></div><div> File "C:\Users\zahransa\anaconda3\lib\site-packages\PIL\Image.py", line 69, in <module><br></div><div> from . import _imaging as core<br></div><div>ImportError: DLL load failed: Le module spécifié est introuvable.<br></div><div><br></div><div><span>Process finished with exit code 1</span><br></div></div><div><br></div><div style="font-family:Calibri,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)"><br></div><div style="font-family:Calibri,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)"><br></div><div style="font-family:Calibri,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">Best regards<br></div><div style="font-family:Calibri,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">Saeed<br></div><div><hr tabindex="-1" style="display:inline-block; width:98%"><br></div><div dir="ltr"><div><span style="font-family:Calibri, sans-serif"><span style="color:#000000"><b>From:</b> mne_analysis-bounces@nmr.mgh.harvard.edu <mne_analysis-bounces@nmr.mgh.harvard.edu> on behalf of mne_analysis-request@nmr.mgh.harvard.edu
<mne_analysis-request@nmr.mgh.harvard.edu><br> <b>Sent:</b> Sunday, April 5, 2020 1:01 AM<br> <b>To:</b> mne_analysis@nmr.mgh.harvard.edu <mne_analysis@nmr.mgh.harvard.edu><br> <b>Subject:</b> Mne_analysis Digest, Vol 147, Issue 12</span></span> </div><div> <br></div></div><div><span style="font-size:13px"><span style="font-size:11pt"><div><div>Send Mne_analysis mailing list submissions to<br></div><div> mne_analysis@nmr.mgh.harvard.edu<br></div><div> <br></div><div> To subscribe or unsubscribe via the World Wide Web, visit<br></div><div> <a href="https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis"> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis</a><br></div><div> or, via email, send a message with subject or body 'help' to<br></div><div> mne_analysis-request@nmr.mgh.harvard.edu<br></div><div> <br></div><div> You can reach the person managing the list at<br></div><div> mne_analysis-owner@nmr.mgh.harvard.edu<br></div><div> <br></div><div> When replying, please edit your Subject line so it is more specific<br></div><div> than "Re: Contents of Mne_analysis digest..."<br></div><div> <br></div><div> <br></div><div> Today's Topics:<br></div><div> <br></div><div> 1. Re: Starting with mne (Dan McCloy)<br></div><div> <br></div><div> <br></div><div> ----------------------------------------------------------------------<br></div><div> <br></div><div> Message: 1<br></div><div> Date: Sat, 04 Apr 2020 22:01:18 +0000<br></div><div> From: Dan McCloy <dan@mccloy.info><br></div><div> Subject: Re: [Mne_analysis] Starting with mne<br></div><div> To: Discussion and support forum for the users of MNE Software<br></div><div> <mne_analysis@nmr.mgh.harvard.edu><br></div><div> Message-ID:<br></div><div> <Mv3TQbne75xhJSyRzIgJ5quw5vT64yrnPlgIilmPnClYppFBc-euEZ-alrg_0xGzNzxz2FkHHpkLgMf-JhvR1lyjJJ_DshC7BidCvKzslew=@mccloy.info><br></div><div> <br></div><div> Content-Type: text/plain; charset="utf-8"<br></div><div> <br></div><div> External Email - Use Caution <br></div><div> <br></div><div> The OPM data are public, and will be automatically downloaded to (by default) $HOME/mne_data/ when you call `mne.datasets.opm.data_path()` for the first time. That function will return the full path to the download location. Can you try running just that line
directly in the python interpreter, rather than running the whole script? There are some reports that URLlib might behave differently in those two conditions (e.g., [here](https://stackoverflow.com/q/27115803/1664024)).<br></div><div> <br></div><div> The other error looks like an error loading matplotlib, and suggests there was some installation problem (specifically with the ft2font dependency). See [here](https://stackoverflow.com/q/24251102/1664024). If I were in this situation I would probably just
start over reinstalling MNE-Python, pay close attention to any warnings / errors that occur during installation, and then copy/paste those warnings/errors into a search engine. If that doesn't lead you to a solution, you can of course ask again here.<br></div><div> <br></div><div> -- dan<br></div><div> Daniel McCloy<br></div><div> <a href="https://dan.mccloy.info">https://dan.mccloy.info</a><br></div><div> Research Scientist<br></div><div> Institute for Learning and Brain Sciences<br></div><div> University of Washington<br></div><div> <br></div><div> ??????? Original Message ???????<br></div><div> On Saturday, April 4, 2020 2:38 PM, Saeed Zahran <saeedzahran@hotmail.com> wrote:<br></div><div> <br></div><div> > External Email - Use Caution<br></div><div> ><br></div><div> > Dear All,<br></div><div> ><br></div><div> > I hope you are fine and in good health,<br></div><div> ><br></div><div> > I am starting a new project, and I would like to use MNE,<br></div><div> > I would like to compare OPM-MEG to SQUID MEG,<br></div><div> > I know mne from past, but this is the first time that I want to install it and use it,<br></div><div> ><br></div><div> > I saw that mne deal with OPM like:<br></div><div> ><br></div><div> > <a href="https://mne.tools/dev/auto_examples/time_frequency/plot_source_power_spectrum_opm.html"> https://mne.tools/dev/auto_examples/time_frequency/plot_source_power_spectrum_opm.html</a><br></div><div> ><br></div><div> > I have some questions and thank you for your answering:<br></div><div> ><br></div><div> > 1) I would like to know if is possible to access the data, for example to look about how the OPM sensors are oriented,<br></div><div> > in the script you put:[subject](https://docs.python.org/3/library/stdtypes.html#str) = 'OPM_sample' so does it mean the sample data are accessable?<br></div><div> > it is possible to get the lead field matrix? sorry if it is simple question because<br></div><div> > I try to install the mne to see what the script should give but when I run it I have some errors, so the second question:<br></div><div> ><br></div><div> > 2) I installed anaconda version 3.7 and pycharm and I follow this link:<br></div><div> > <a href="https://mne.tools/stable/install/mne_python.html">https://mne.tools/stable/install/mne_python.html</a><br></div><div> ><br></div><div> > I run an example from mne, I get the below error:<br></div><div> ><br></div><div> > C:\Users\zahransa\anaconda3\python.exe C:/Users/zahransa/PycharmProjects/opm/plot_opm_data.py<br></div><div> > Using default location ~/mne_data for opm...<br></div><div> > Downloading archive MNE-OPM-data.tar.gz to C:\Users\zahransa\mne_data<br></div><div> > Error while fetching file <a href="https://osf.io/p6ae7/download?version=2">https://osf.io/p6ae7/download?version=2</a>. Dataset fetching aborted.<br></div><div> > Traceback (most recent call last):<br></div><div> > File "C:/Users/zahransa/PycharmProjects/opm/plot_opm_data.py", line 20, in <module><br></div><div> > data_path = mne.datasets.opm.data_path()<br></div><div> > File "<decorator-gen-411>", line 21, in data_path<br></div><div> > File "C:\Users\zahransa\anaconda3\lib\site-packages\mne\datasets\opm\opm.py", line 21, in data_path<br></div><div> > download=download)<br></div><div> > File "C:\Users\zahransa\anaconda3\lib\site-packages\mne\datasets\utils.py", line 395, in _data_path<br></div><div> > remove_archive, full = _download(path, u, an, h)<br></div><div> > File "C:\Users\zahransa\anaconda3\lib\site-packages\mne\datasets\utils.py", line 453, in _download<br></div><div> > hash_=hash_, hash_type=hash_type)<br></div><div> > File "<decorator-gen-3>", line 21, in _fetch_file<br></div><div> > File "C:\Users\zahransa\anaconda3\lib\site-packages\mne\utils\fetching.py", line 117, in _fetch_file<br></div><div> > _get_http(url, temp_file_name, initial_size, timeout, verbose_bool)<br></div><div> > File "C:\Users\zahransa\anaconda3\lib\site-packages\mne\utils\fetching.py", line 43, in _get_http<br></div><div> > response = request.urlopen(request.Request(url), timeout=timeout)<br></div><div> > File "C:\Users\zahransa\anaconda3\lib\urllib\request.py", line 222, in urlopen<br></div><div> > return opener.open(url, data, timeout)<br></div><div> > File "C:\Users\zahransa\anaconda3\lib\urllib\request.py", line 525, in open<br></div><div> > response = self._open(req, data)<br></div><div> > File "C:\Users\zahransa\anaconda3\lib\urllib\request.py", line 548, in _open<br></div><div> > 'unknown_open', req)<br></div><div> > File "C:\Users\zahransa\anaconda3\lib\urllib\request.py", line 503, in _call_chain<br></div><div> > result = func(*args)<br></div><div> > File "C:\Users\zahransa\anaconda3\lib\urllib\request.py", line 1389, in unknown_open<br></div><div> > raise URLError('unknown url type: %s' % type)<br></div><div> > urllib.error.URLError: <urlopen error unknown url type: https><br></div><div> ><br></div><div> > another example:<br></div><div> ><br></div><div> > C:\Users\zahransa\anaconda3\python.exe C:/Users/zahransa/PycharmProjects/opm/plot_mne_inverse_connectivity_spectrum.py<br></div><div> > Traceback (most recent call last):<br></div><div> > File "C:/Users/zahransa/PycharmProjects/opm/plot_mne_inverse_connectivity_spectrum.py", line 13, in <module><br></div><div> > import matplotlib.pyplot as plt<br></div><div> > File "C:\Users\zahransa\anaconda3\lib\site-packages\matplotlib\__init__.py", line 205, in <module><br></div><div> > _check_versions()<br></div><div> > File "C:\Users\zahransa\anaconda3\lib\site-packages\matplotlib\__init__.py", line 190, in _check_versions<br></div><div> > from . import ft2font<br></div><div> > ImportError: DLL load failed: Le module sp?cifi? est introuvable.<br></div><div> ><br></div><div> > Thank you for your help,<br></div><div> ><br></div><div> > Best regards<br></div><div> > Saeed Zahran<br></div><div> ><br></div><div> > ---------------------------------------------------------------<br></div><div> ><br></div><div> > From: mne_analysis-bounces@nmr.mgh.harvard.edu <mne_analysis-bounces@nmr.mgh.harvard.edu> on behalf of mne_analysis-request@nmr.mgh.harvard.edu <mne_analysis-request@nmr.mgh.harvard.edu><br></div><div> > Sent: Saturday, April 4, 2020 12:35 PM<br></div><div> > To: mne_analysis@nmr.mgh.harvard.edu <mne_analysis@nmr.mgh.harvard.edu><br></div><div> > Subject: Mne_analysis Digest, Vol 147, Issue 10<br></div><div> ><br></div><div> > Send Mne_analysis mailing list submissions to<br></div><div> > mne_analysis@nmr.mgh.harvard.edu<br></div><div> ><br></div><div> > To subscribe or unsubscribe via the World Wide Web, visit<br></div><div> > <a href="https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis"> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis</a><br></div><div> > or, via email, send a message with subject or body 'help' to<br></div><div> > mne_analysis-request@nmr.mgh.harvard.edu<br></div><div> ><br></div><div> > You can reach the person managing the list at<br></div><div> > mne_analysis-owner@nmr.mgh.harvard.edu<br></div><div> ><br></div><div> > When replying, please edit your Subject line so it is more specific<br></div><div> > than "Re: Contents of Mne_analysis digest..."<br></div><div> ><br></div><div> > Today's Topics:<br></div><div> ><br></div><div> > 1. Re: Temporal Generalization - Different results with and<br></div><div> > without using cross validation (Maryam Zolfaghar)<br></div><div> > 2. MNE-BIDS 0.4 released! (Stefan Appelhoff)<br></div><div> ><br></div><div> > ----------------------------------------------------------------------<br></div><div> ><br></div><div> > Message: 1<br></div><div> > Date: Fri, 3 Apr 2020 16:17:56 -0400<br></div><div> > From: Maryam Zolfaghar <Maryam.Zolfaghar@colorado.edu><br></div><div> > Subject: Re: [Mne_analysis] Temporal Generalization - Different<br></div><div> > results with and without using cross validation<br></div><div> > To: Alexandre Gramfort <alexandre.gramfort@inria.fr><br></div><div> > Cc: Discussion and support forum for the users of MNE Software<br></div><div> > <mne_analysis@nmr.mgh.harvard.edu><br></div><div> > Message-ID:<br></div><div> > <CAJOF5UAMPtJZPd-_CrmK8=GfZJ34Sm9m=5-weidvTbmBT_+vxw@mail.gmail.com><br></div><div> > Content-Type: text/plain; charset="utf-8"<br></div><div> ><br></div><div> > External Email - Use Caution<br></div><div> ><br></div><div> > Hi Alex,<br></div><div> ><br></div><div> > Thanks a lot for the helpful information. I have run my code according to<br></div><div> > your suggestion but I want to make sure I understood your point correctly.<br></div><div> ><br></div><div> > As you suggested, the difference in results might be due to "shuffle=True".<br></div><div> > However, I do not understand why you suggested "*StratifiedShuffleSplit* "?<br></div><div> > what would be the difference between doing "cv = *StratifiedKFold*<br></div><div> > *(n_splits=5*, shuffle=True, random_state=42) or cv =<br></div><div> > *StratifiedKFold**(*n_splits=5,<br></div><div> > *shuffle=False*)" vs "cv = *StratifiedShuffleSplit*(*n_splits=1000*,<br></div><div> > random_state=42)"? Why not only setting random state and just using "cv =<br></div><div> > *StratifiedShuffleSplit*(*n_splits=5*, random_state=42)" with the same<br></div><div> > number of splits?<br></div><div> ><br></div><div> > Thanks,<br></div><div> > -Maryam<br></div><div> ><br></div><div> > On Wed, Mar 25, 2020 at 4:15 PM Alexandre Gramfort <<br></div><div> > alexandre.gramfort@inria.fr> wrote:<br></div><div> ><br></div><div> >> hi,<br></div><div> >><br></div><div> >> when you do:<br></div><div> >><br></div><div> >> *cv = StratifiedKFold(n_splits=5, shuffle=True)*<br></div><div> >><br></div><div> >> *you make the cross-validation folds random (shuffle=True). It means*<br></div><div> >> *that everytime you run the code you will get a different value. This the<br></div><div> >> inherent*<br></div><div> >><br></div><div> >> *variance of the statistic that cross-validation reports.*<br></div><div> >><br></div><div> >> *To avoid this randomness (which should not be neglected) you can fix the<br></div><div> >> random state eg*<br></div><div> >><br></div><div> >> cv = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)<br></div><div> >><br></div><div> >> to avoid this you should use a StratifiedShuffleSplit and use many folds<br></div><div> >> eg 100 to limit<br></div><div> >> the variance due to the choice of the cv splits.<br></div><div> >><br></div><div> >> HTH<br></div><div> >> Alex<br></div><div> >><br></div><div> >><br></div><div> >><br></div><div> >> On Wed, Mar 25, 2020 at 4:39 PM Maryam Zolfaghar <<br></div><div> >> Maryam.Zolfaghar@colorado.edu> wrote:<br></div><div> >><br></div><div> >>> External Email - Use Caution<br></div><div> >>><br></div><div> >>> Hi MNE experts,<br></div><div> >>><br></div><div> >>> I am using the temporal generalization<br></div><div> >>> <<a href="https://secure-web.cisco.com/1jWVu354v-yk2PHG6scpRuNlqi-Xei3wH-moV9-Av0fstdMSB7aUnzzNvmfnF3JuIo6EC0A07IdflTTdP4WpUqpUrZSclzoSV8WBksahu0ODpDgFOsHmm6eHO0i8QuTC6bAvL9yb4XuOmHH4wqg_ubbDuEIZUQriZvysI-ayKjx5o4jndHXz4UqLxv0CVaHv0YafoA8c9X29SwV52DKgOaPlf2v1oiEOTcd-Wu6QtvRMnn5AgMVc4_QxbHaXjM5qAZz8u1r_wFixIzhsSgFa89d05jmOT0BVNWMZ1-sXCPt0by2krKKqiEyV7OLYBLLMrTTnWZ00KZGIGX_VkqbIA4uArHPJxE7w_U3mNvn0YFXUUhVuopodRcQf_ZA81GJA5E5Qy5s6PW0ZP8SjUtICoajBkEHmm6LpIKycpscbg0W7uVdtFNyDkDTMXVq0OemdyrFxvdqDi6VnKPtgefbyyNw/https%3A%2F%2Fmne.tools%2Fdev%2Fauto_tutorials%2Fmachine-learning%2Fplot_sensors_decoding.html%23temporal-generalization">https://secure-web.cisco.com/1jWVu354v-yk2PHG6scpRuNlqi-Xei3wH-moV9-Av0fstdMSB7aUnzzNvmfnF3JuIo6EC0A07IdflTTdP4WpUqpUrZSclzoSV8WBksahu0ODpDgFOsHmm6eHO0i8QuTC6bAvL9yb4XuOmHH4wqg_ubbDuEIZUQriZvysI-ayKjx5o4jndHXz4UqLxv0CVaHv0YafoA8c9X29SwV52DKgOaPlf2v1oiEOTcd-Wu6QtvRMnn5AgMVc4_QxbHaXjM5qAZz8u1r_wFixIzhsSgFa89d05jmOT0BVNWMZ1-sXCPt0by2krKKqiEyV7OLYBLLMrTTnWZ00KZGIGX_VkqbIA4uArHPJxE7w_U3mNvn0YFXUUhVuopodRcQf_ZA81GJA5E5Qy5s6PW0ZP8SjUtICoajBkEHmm6LpIKycpscbg0W7uVdtFNyDkDTMXVq0OemdyrFxvdqDi6VnKPtgefbyyNw/https%3A%2F%2Fmne.tools%2Fdev%2Fauto_tutorials%2Fmachine-learning%2Fplot_sensors_decoding.html%23temporal-generalization</a>>
approach.<br></div><div> >>> I have plotted scores' output from cross_val_multiscore<br></div><div> >>> <<a href="https://secure-web.cisco.com/1h1fBYrY82nc7Ha3ohR5V_l8-7FgWv1EKHCEhN_hAJm8A-NfF2V7uxTWmy5IwO-cOk33FiOpkywzhHZcxjWQfQez3GPysxi2nGr5NkcO-UbypLFGD_4d5yopqahopnYqwDy-b5yOA56uFYbRzHJLYyu6NsGNICQkbBWbwDt1EXOgxAFYg6wK5WIcxNYuZ9Jg_4G88dlreZRD8p-yNwFZ7D7z_DBKIo8eudHagdQhME0woFuPo63yrIoM4zLcAWLhFta-xNl2IvYh9iTIk5U5zUwFIbyRBa4HIWRD-Es3MGFPRcDmjke4u-MSG4yKuwjcIIfy3K4MCoiMF79wTrt2u-SutYTmMNte1IFDKYhw-kx7oH-XbtFkLZH1xigiLkdICoIFeZnptCIYr0Cu1mb75-E8SZD17l2y3xvLlgqZQ2aOYCPoFgCdlbtaBN7GhzUzeunr0EMGN7y6_W-4o-9ASig/https%3A%2F%2Fmne.tools%2Fdev%2Fgenerated%2Fmne.decoding.cross_val_multiscore.html%23mne.decoding.cross_val_multiscore">https://secure-web.cisco.com/1h1fBYrY82nc7Ha3ohR5V_l8-7FgWv1EKHCEhN_hAJm8A-NfF2V7uxTWmy5IwO-cOk33FiOpkywzhHZcxjWQfQez3GPysxi2nGr5NkcO-UbypLFGD_4d5yopqahopnYqwDy-b5yOA56uFYbRzHJLYyu6NsGNICQkbBWbwDt1EXOgxAFYg6wK5WIcxNYuZ9Jg_4G88dlreZRD8p-yNwFZ7D7z_DBKIo8eudHagdQhME0woFuPo63yrIoM4zLcAWLhFta-xNl2IvYh9iTIk5U5zUwFIbyRBa4HIWRD-Es3MGFPRcDmjke4u-MSG4yKuwjcIIfy3K4MCoiMF79wTrt2u-SutYTmMNte1IFDKYhw-kx7oH-XbtFkLZH1xigiLkdICoIFeZnptCIYr0Cu1mb75-E8SZD17l2y3xvLlgqZQ2aOYCPoFgCdlbtaBN7GhzUzeunr0EMGN7y6_W-4o-9ASig/https%3A%2F%2Fmne.tools%2Fdev%2Fgenerated%2Fmne.decoding.cross_val_multiscore.html%23mne.decoding.cross_val_multiscore</a>><br></div><div> >>> with a RepeatedStratifiedKFold<br></div><div> >>> <<a href="https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.RepeatedStratifiedKFold.html#sklearn.model_selection.RepeatedStratifiedKFold">https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.RepeatedStratifiedKFold.html#sklearn.model_selection.RepeatedStratifiedKFold</a>><br></div><div> >>> cv parameter. I have also plotted training scores<br></div><div> >>><br></div><div> >>> I assumed that I should get the same result and the only difference would<br></div><div> >>> be the diagonal results where the diagonal training scores will be all 1.<br></div><div> >>> However, the general results are quite different (you can still see some<br></div><div> >>> fade underlying pattern similar in both).<br></div><div> >>><br></div><div> >>> Any idea of why plotting scores using cross-validation v.s. only<br></div><div> >>> plotting fitting/training scores will give different results?<br></div><div> >>><br></div><div> >>> This is my understanding of what should be going on: in the training case<br></div><div> >>> without using any cross-validation, on each time point, there was a<br></div><div> >>> classifier/decoder that was trained by seeing all EEG channels' data over<br></div><div> >>> all epochs at that train time point, therefore it would give a perfect<br></div><div> >>> score on the same test time point. However, a different time point<br></div><div> >>> (testing times) has different data that can be seen as a test set for this<br></div><div> >>> decoder. Right? (Even if there was an autocorrelation between EEG data over<br></div><div> >>> time and still see some meaningful pattern in time generalization matrix,<br></div><div> >>> it means that EEG data had task-related information over time which is<br></div><div> >>> still meaningful).<br></div><div> >>><br></div><div> >>> ---------<br></div><div> >>> I have also put my code here:<br></div><div> >>><br></div><div> >>> *Scores using cross-validation:*<br></div><div> >>><br></div><div> >>> *clf_SVC = make_pipeline(*<br></div><div> >>> * StandardScaler(),*<br></div><div> >>> * LinearModel(LinearSVC(random_state=0,<br></div><div> >>> max_iter=10000)))*<br></div><div> >>><br></div><div> >>> *temp_gen = GeneralizingEstimator(clf_SVC, scoring='roc_auc',<br></div><div> >>> n_jobs=1,verbose=True)*<br></div><div> >>><br></div><div> >>> *cv = StratifiedKFold(n_splits=5, shuffle=True)*<br></div><div> >>> *scores = cross_val_multiscore(temp_gen, X, y, cv=cv, n_jobs=1)*<br></div><div> >>><br></div><div> >>> *Only fitting scores:*<br></div><div> >>><br></div><div> >>> *temp_gen.fit(X=X ,y=y)*<br></div><div> >>> *scores = temp_gen.score(X=X, y=y) #scores without cv*<br></div><div> >>> *-----------*<br></div><div> >>><br></div><div> >>> - I will appreciate any comments,<br></div><div> >>> Thanks<br></div><div> >>><br></div><div> >>><br></div><div> >>><br></div><div> >>><br></div><div> >>><br></div><div> >>><br></div><div> >>><br></div><div> >>><br></div><div> >>> _______________________________________________<br></div><div> >>> Mne_analysis mailing list<br></div><div> >>> Mne_analysis@nmr.mgh.harvard.edu<br></div><div> >>> <a href="https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis">https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis</a><br></div><div> >><br></div><div> >><br></div><div> > -------------- next part --------------<br></div><div> > An HTML attachment was scrubbed...<br></div><div> > URL: <a href="http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20200403/556f65b1/attachment-0001.html"> http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20200403/556f65b1/attachment-0001.html</a><br></div><div> ><br></div><div> > ------------------------------<br></div><div> ><br></div><div> > Message: 2<br></div><div> > Date: Sat, 4 Apr 2020 11:35:13 +0200<br></div><div> > From: Stefan Appelhoff <stefan.appelhoff@mailbox.org><br></div><div> > Subject: [Mne_analysis] MNE-BIDS 0.4 released!<br></div><div> > To: mne_analysis@nmr.mgh.harvard.edu<br></div><div> > Message-ID: <72db9be7-7537-77b9-5685-3a3a0d8b1d7b@mailbox.org><br></div><div> > Content-Type: text/plain; charset="utf-8"; format="flowed"<br></div><div> ><br></div><div> > External Email - Use Caution<br></div><div> ><br></div><div> > Dear MNE community,<br></div><div> ><br></div><div> > shortly after MNE 0.20 was released, we now proudly release MNE-BIDS 0.4<br></div><div> > for all of you using the Brain Imaging Data Structure (BIDS) in your work.<br></div><div> ><br></div><div> > Next to several bug fixes and an improved documentation, there are some<br></div><div> > new exciting features such as an improved automatic anonymization of<br></div><div> > datasets. Check it out!<br></div><div> ><br></div><div> > - BIDS: <a href="https://bids.neuroimaging.io/">https://bids.neuroimaging.io/</a><br></div><div> ><br></div><div> > - MNE-BIDS: <a href="https://mne.tools/mne-bids/stable/index.html">https://mne.tools/mne-bids/stable/index.html</a><br></div><div> ><br></div><div> > Best regards,<br></div><div> ><br></div><div> > your MNE-BIDS team<br></div><div> ><br></div><div> > ------------------------------<br></div><div> ><br></div><div> > _______________________________________________<br></div><div> > Mne_analysis mailing list<br></div><div> > Mne_analysis@nmr.mgh.harvard.edu<br></div><div> > <a href="https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis">https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis</a><br></div><div> ><br></div><div> > End of Mne_analysis Digest, Vol 147, Issue 10<br></div><div> > *********************************************<br></div><div> -------------- next part --------------<br></div><div> An HTML attachment was scrubbed...<br></div><div> URL: <a href="http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20200404/021bcd69/attachment.html"> http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20200404/021bcd69/attachment.html</a> <br></div><div> <br></div><div> ------------------------------<br></div><div> <br></div><div> _______________________________________________<br></div><div> Mne_analysis mailing list<br></div><div> Mne_analysis@nmr.mgh.harvard.edu<br></div><div> <a href="https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis">https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis</a><br></div><div> <br></div><div> End of Mne_analysis Digest, Vol 147, Issue 12<br></div><div> *********************************************<br></div></div></span></span></div></blockquote><div><br></div>