A user writes...

> 2. I was trying svdpackout.pl on the output of las2 with the apw.100000
> corpus. svdpackout.pl kept running for a really long time. Is this
> expected? Would you have any idea how long svdpackout could take on a
> corpus with 100000 sentences?

The more crucial question is how large was the input matrix that went to
svd? Depending on that size, it might take a long time. In effect,
svdpackout is trying to regenerate the input matrix, now with the svd
processed cell values. That means that most of the cells will have a
non-zero value, and you will have a huge matrix to write. So, what is
often recommended is that you only output the Mxk reduced matrix, since
this will give you exactly the information you need or want for making
comparisons.

In other words, suppose you have an MxN input matrix, that you reduce
to k dimensions with SVD. svdpackout will output an MxN matrix with
"smoothed" values. However, you can output the Mxk matrix if you use
the --rowonly option, and that will run much much faster.

--
Ted Pedersen
http://www.d.umn.edu/~tpederse


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now. 
http://productguide.itmanagersjournal.com/
_______________________________________________
senseclusters-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/senseclusters-users

Reply via email to