Dear all,
I would like to fit some data (few samples and a much larger number of
features), using linear regression with lasso regularization.
So far I have used the packages enet and lars, that perform very well and
of which I really appreciated the easy interface.
Unfortunately, while on the majority of my examples the work well, on some
of my datasets the training fail.
I contacted the authors who explained me that there is some issue due to
the Cholesky decomposition, and numerical precision, and they have been
kind enough to explain me some workarounds (involving modifying two of the
parameters).
While this solve the issues in some cases, in other I am still unable to
train a regressor.
I was wondering if you know of any other reliable package that I could use
to build a linear model using lasso regularization, so that I could find a
linear model for the other cases I am missing.
Thanks a lot in advance for your help,
Cheers,
Luca
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
