Error in downloading: " urlopen error [Errno 60] Operation timed out "

[Jeansma]

Hi,
I tried to run the "download_abide_preproc.py" script to download ABIDE preprocessed data from S3, but met the following error:
urllib.error.URLError: <urlopen error [Errno 60] Operation timed out>
I am wondering if the given url is still working? Or is there any update for that?

Any suggestion or comment is much appreciated!

Thank you!
Guixiang

[anibalsolon]

Hi Guixiang, Yes, it is still working! I've just tested here on my laptop. Could you please provide us with the arguments that you are using to download the ABIDE dataset? On the download_abide_preproc.py script. This combination of arguments worked for me: ```download_abide_preproc.py -d reho -p cpac -s filtnoglobal -o reho -t PITT``` It could also be an instability on your network. Thank you for reaching out! Anibal Sólon Em seg, 18 de jun de 2018 às 13:39, Guixiang Ma <notifications@github.com> escreveu:

…

Hi, I tried to run the "download_abide_preproc.py" script to download ABIDE preprocessed data from S3, but met the following error: urllib.error.URLError: <urlopen error [Errno 60] Operation timed out> I am wondering if the given url is still working? Or is there any update for that? Any suggestion or comment is much appreciated! Thank you! Guixiang — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#15>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAiVXbJH6UdEVGJtOArVi1j6X68Ufdl_ks5t9-XWgaJpZM4UsL_J> .

[AbdelbassetBrahim]

Hi,
I used the "download_abide_preproc.py" script to download ABIDE preprocessed, as following, for CMU site: python download_abide_preproc_up.py -d func_preproc -p cpac -s filt_global -o /path for data -t CMU
However the resulting files are not for all the subjects in the site and there are several error messages:"Error extracting info from phenotypic file, skipping..."
I tried for other sites also and the number of downloaded files always less than the number of subjects.
I used also the fetch_abide_pcp function in nilearn and I did not get all the subjects for each site.
Could you please provide me a script to download all the subjects for each site.
Thank you in advance.

[anibalsolon]

Hi Abdelbasset, Have you modified the script by any change? I noticed that it has an "_up" in the file name. In any case, the package *nilearn* provides a function to download the ABIDE dataset. For now, it does not offer you an option to select data from a specific site, but I will open a pull request to change it. https://nilearn.github.io/modules/generated/nilearn.datasets.fetch_abide_pcp.html Best, Anibal Solon Assistant Research Scientist Computational Neuroimaging Laboratory p: 646.625.4378 e: anibal.solon@childmind.org <steven.giavasis@childmind.org> Child Mind Institute 445 Park Avenue (entrance on 56th Street) New York, NY 10022 childmind.org <https://aidbox.org/ct/?i=638132000000914&t=1>

…

On Fri, Jan 11, 2019 at 11:15 AM AbdelbassetBrahim ***@***.***> wrote: Hi, I used the "download_abide_preproc.py" script to download ABIDE preprocessed, as following, for CMU site: python download_abide_preproc_up.py -d func_preproc -p cpac -s filt_global -o /path for data -t CMU However the resulting files are not for all the subjects in the site and there are several error messages:"Error extracting info from phenotypic file, skipping..." I tried for other sites also and the number of downloaded files always less than the number of subjects. I used also the fetch_abide_pcp function in nilearn and I did not get all the subjects for each site. Could you please provide me a script to download all the subjects for each site. Thank you in advance. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#15 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAiVXQXriBewbdOjlKhz0PkSKtGXjsgPks5vCLi9gaJpZM4UsL_J> .

[anibalsolon]

Just found out that they already have the site filter, but it is not clear on the documentation.

For example, this python command will download your data into "/tmp":

nilearn.datasets.fetch_abide_pcp('/tmp',
                                 derivatives=['func_preproc'],
                                 pipeline='cpac',
                                 band_pass_filtering=True,
                                 global_signal_regression=True,
                                 SITE_ID=['NYU'])

[AbdelbassetBrahim]

Hi Anibal,
Thank you for your replies. Yes I used at the end the fetch_abide_pcp function from nilearn to download the preprocessed data of Abide I according to each site but always the number of downloaded data (preprocessed by capc) for the subjects is less than the number of subjects presented in this link:
http://fcon_1000.projects.nitrc.org/indi/abide/abide_I.html
for instance, I got 871 preprocessed data for all the subjects however the number of subjects for all the sites is more than 1000 subjects.

[anibalsolon]

It is probably because of the quality control.

You can disable the quality control with the option quality_checked=False.

[Devika-k]

@Jeansma @anibalsolon @AbdelbassetBrahim @czarrar @pbellec

Recently I have downloaded ABIDE 1 Preprocessed dataset by fetching it from “Nilearn”.
pipeline: cpac
derivatives: func_preproc

The code is here

abide_dataset = nilearn.datasets.fetch_abide_pcp(data_dir=’/content/ABIDE’, n_subjects=None, pipeline=‘cpac’, band_pass_filtering=True, global_signal_regression=True, derivatives=[‘func_preproc’], quality_checked=False, verbose=1)

By disabling the quality control parameter I was able to download fMRI images of only 1035 subjects… In the phenotype information (CSV file) remaining 77 subjects file_name was mentioned as no_filename…

Does these subjects have only phenotype information (no fMRI images)?
Does the number of subject’s (fMRI images) depends on the choice of pipeline and derivative?
Is it possible to download fMRI images of 1112 subjects using nilearn?

It will be of great help if any one can give a clarification

[AbdelbassetBrahim]

Hi,accoding to my experience with this database and when using nilearn.datasets.fetch_abide_pcp, you can download till 1035 subjects when quality_checked=False. However, you will have 871 subjects when quality_checked=True. This is the only input parameter in this function that can affects the number of subjects...Hope it helps.Cordially. Le jeudi 3 octobre 2019 à 11:19:39 UTC+2, Devika-k <notifications@github.com> a écrit : @Jeansma @anibalsolon @AbdelbassetBrahim @czarrar @pbellec Recently I have downloaded ABIDE 1 Preprocessed dataset by fetching it from “Nilearn”. pipeline: cpac derivatives: func_preproc The code is here abide_dataset = nilearn.datasets.fetch_abide_pcp(data_dir=’/content/ABIDE’, n_subjects=None, pipeline=‘cpac’, band_pass_filtering=True, global_signal_regression=True, derivatives=[‘func_preproc’], quality_checked=False, verbose=1) By disabling the quality control parameter I was able to download fMRI images of only 1035 subjects… In the phenotype information (CSV file) remaining 77 subjects file_name was mentioned as no_filename… Does these subjects have only phenotype information (no fMRI images)? Does the number of subject’s (fMRI images) depends on the choice of pipeline and derivative? Is it possible to download fMRI images of 1112 subjects using nilearn? It will be of great help if any one can give a clarification — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

[Devika-k]

@AbdelbassetBrahim
Thanks for your reply..
Is it possible to download fMRI images of 1112 subjects other than using nilearn?

[AbdelbassetBrahim]

You can try this python script and see: download_abide_preproc.py

[Pranjali87]

hi, while downloading abide dataset using fetch_abide_pcp , I am getting connection error in between .
after PITT, in OLIN id I am getting error.
I tried 3 times but every time I got connection error from host.
can anybody please help me ?