Hi and thank you for your reply!

I will try the text2sval2.pl option by marking the lemma in the first
column with <head>noun</head> and removing the freq information in the 4th
column by using the non-sorted file. In this way, if I have read correctly
the description of the input required for the preprocess, I should have the
one-line per context requirement fulfilled, correct? In this case, after
processing with texts2sval2.pl, I can use one of the vector convertors for
the proper input for clusterstopping.pl?

In regards to the sample program that I was trying. I wanted to get a feel
for what the results would look at. In this way, I copied the example input
in sparse matrix format (from the clusterstopping.pl documentation page)
and I simply ran the command:
* $perl clusterstopping.pl <http://clusterstopping.pl> sample_input.txt*

As the arguments are optional, I just wanted to understand what the results
would look like before implementing any task-specific parameters and the
error mentioned in my original email occurred.

Thank you again, in advance, for any insight to understand better both the
program and how to solve my error!!
Lauren






On Sat, Nov 9, 2013 at 8:18 PM, Ted Pedersen <[email protected]> wrote:

> Perhaps the easiest way to put data into the format required by
> SenseClusters is by using one of the converter programs we have. In
> your case I think text2sval.pl would be the right choice...
>
> It you have SenseClusters installed, you should simply be able to run
>
> text2sval2.pl
>
> (there are a few options you could use, all of which is described
> below, or you could run
>
> text2sval.pl --help
>
>
> http://cpansearch.perl.org/src/TPEDERSE/Text-SenseClusters-1.03/Toolkit/preprocess/plain/text2sval.pl
>
> About the sample program you are trying to run and getting the error
> with, can you let me know the command you were running?
>
> Thanks!
> Ted
>
> On Sat, Nov 9, 2013 at 9:46 AM, Lauren Romeo <[email protected]>
> wrote:
> > Hi Professor Pederson,
> >
> > I am a new SenseClusters user and I am particularly interested in one
> Tool
> > :: ClusterStopping. I have been trying to implement it using it as a
> > standalone part of an experiment that I am running.
> >
> > In this way,  I already have data that I am working with (3-column
> > tab-separated format (target, slot-filler, weight)).
> >
> > Here is  small (10-line) sample of the format of my input:
> > abduction-n    into+n-the+n-a-j-loss-n    1
> > abduction-n    into+n-the+n-a-small-cut-n    2
> > abduction-n    into+n-the+n-j-bleeding-n    1
> > abduction-n    into+n-the+n-j-loss-n    1
> > zoonosis-n    of+n-j+n-the-location-n    1
> > zoonosis-n    of+n-j+n-the-world-n    1
> > zoonosis-n    of+n-j+n-the-development-n    1
> > zoonosis-n    of+n-j+n-the-j-collection-n    1
> > zoonosis-n    of+n-j+n-the-j-success-n    1
> > zoonosis-n    of+n-j+ns-photo-n    1
> >
> >
> > I have unsuccessfully been able to determine how I can translate my data
> to
> > a usable input file for this particular process.
> >
> > It is a rather large file (5GB). I am also not clear as to 1. if there
> is an
> > option to convert directly this format of
> > data into a sparse format (considering the size I suppose that is the
> best
> > option), however, the sample data that you provide in the website (I
> used to
> > understand what I am
> > doing) - continuously gives me the following error -- which is why I have
> > been unable to solve the problem on my own --
> > it is because I do not understand what the machine is requiring from me.
> >
> > sh: cannot create
> > /root/.cpan/build/Text-
> >
> SenseClusters-1.03-5f8DVC/Toolkit/clusterstop//homedtic/usr/SenseCluster/sense_clusters-sample.rtf.1:
> > Directory nonexistent
> > Error while running vcluster --clmethod rb --crfun i2 --sim
> > cos --rowmodel none --colmodel none --nooutput
> > /homedtic/usr/SenseCluster/sense_clusters-sample.rtf 1
> >
> >
> > Any assistance that you can provide to me would be fantastic, I have been
> > really trying to search online for
> > answers --- looking at the sample testdata in available in the program
> etc.,
> > but I have come to a deadend. Do you think
> > you would be able to provide me with any assistance regarding how my data
> > can be potentially used with your
> > clusterstopping.pl program?
> >
> > In advance, thank you very much for any assistance you might
> > be able to give.
> >
> > Again, thank you very much.
> >
> >
> >
> >
> ------------------------------------------------------------------------------
> > November Webinars for C, C++, Fortran Developers
> > Accelerate application performance with scalable programming models.
> Explore
> > techniques for threading, error checking, porting, and tuning. Get the
> most
> > from the latest Intel processors and coprocessors. See abstracts and
> > register
> >
> http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk
> > _______________________________________________
> > senseclusters-users mailing list
> > [email protected]
> > https://lists.sourceforge.net/lists/listinfo/senseclusters-users
> >
>
>
>
> --
> Ted Pedersen
> http://www.d.umn.edu/~tpederse
>
>
> ------------------------------------------------------------------------------
> November Webinars for C, C++, Fortran Developers
> Accelerate application performance with scalable programming models.
> Explore
> techniques for threading, error checking, porting, and tuning. Get the most
> from the latest Intel processors and coprocessors. See abstracts and
> register
> http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk
> _______________________________________________
> senseclusters-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/senseclusters-users
>



-- 
Lauren Romeo
------------------------------------------------------------------------------
November Webinars for C, C++, Fortran Developers
Accelerate application performance with scalable programming models. Explore
techniques for threading, error checking, porting, and tuning. Get the most 
from the latest Intel processors and coprocessors. See abstracts and register
http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk
_______________________________________________
senseclusters-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/senseclusters-users

Reply via email to