UDPIPE not working

[rjzevallos]

Someone uses UDPIPE, the service is down. Y_Y

link service: http://lindat.mff.cuni.cz/services/udpipe/api/process

[matyaskopp]

I don't see any trouble. It works. You can always check the web frontend: https://lindat.mff.cuni.cz/services/udpipe/

If too many users use it, it can be overloaded, and the processing is done on CPUs which are much slower than GPUs.

[matyaskopp]

You can also use my perl script for annotations if you don't want to implement it yourself: https://github.com/ufal/ParCzech/tree/master/src/udpipe2

[rjzevallos]

When I run

'perl -I lib udpipe2/udpipe2.pl'

I get:

Can't locate XML/LibXML/PrettyPrint.pm in @INC (you may need to install the XML::LibXML::PrettyPrint module) (@INC contains: lib /etc/perl /usr/local/lib/x86_64-linux-gnu/perl/5.30.0 /usr/local/share/perl/5.30.0 /usr/lib/x86_64-linux-gnu/perl5/5.30 /usr/share/perl5 /usr/lib/x86_64-linux-gnu/perl/5.30 /usr/share/perl/5.30 /usr/local/lib/site_perl /usr/lib/x86_64-linux-gnu/perl-base) at lib/ParCzech/PipeLine/FileManager.pm line 405. BEGIN failed--compilation aborted at lib/ParCzech/PipeLine/FileManager.pm line 405. Compilation failed in require at udpipe2/udpipe2.pl line 16. BEGIN failed--compilation aborted at udpipe2/udpipe2.pl line 16.

[matyaskopp]

Yes, because you don't have installed dependencies.

cpan XML::LibXML::PrettyPrint

Then try to run again and install the next missing dependency.

For XML:LibXML you will probably need to install some system dependency (in Ubuntu-like systems libxml2-dev)

[rjzevallos]

Done! Thank you so much. I would like get this using your code:

curl -F input=vertical -F data=@tokens_vertical.txt -F model=catalan-ancora-ud-2.10-220711 -F parser= http://lindat.mff.cuni.cz/services/udpipe/api/process > output_ud.json"

I was trying put this but is not the same:

perl -I lib udpipe2/udpipe2.pl --colon2underscore \ --model "catalan-ancora-ud-2.10-220711" \ --elements "seg" \ --debug \ --try2continue-on-error \ --filelist list_of_filenames2process.fl \ --input-dir inDir \ --output-dir outDir

[matyaskopp]

My code annotates TEI files, so the input is TEI files, and the output is annotated TEI files.

BTW, why are we discussing this? We already have an ES-CT sample: https://github.com/clarin-eric/ParlaMint/tree/data/Data/ParlaMint-ES-CT so I guess you have a working pipeline...

[TomazErjavec]

My code annotates TEI files, so the input is TEI files, and the output is annotated TEI files.

But in a very general sense it would be great if anybody (for a certain value of anybody) could make a TEI file and produce with this script a .TEI.ana without having to have their own pipeline. It would take "only" writing a section in contributing?

[TomazErjavec]

My suggestion probably won't happen, so I am closing this issue. If anyone disagrees, pls. reopen.