R help - Sep 2008 - persistence of model in R to a file

Hi,
    Is there a way to save R models (glm, lm , rpart etc) in a file that
be read in later?
 
    I noticed models take up space. by space them off and removing them
from memory it seems that would be useful.
 
    Also why do the models keep a copy of all columns in the original
data set even those columns are not in the model.
    E.g. if I build a model on columns A, B even thought column C exists
in the dataframe when I run the model against test data with only
columns A and B then I get an error saying column C is missing from the
data.
 
    Anyway it seems like the models would be more compact without having
to include within them data from the original data set used to build
them but that is an aside.
 
    If models can be persisted and loaded selectively later that would
neat.
 
 
thanks
Dhruv

	[[alternative HTML version deleted]]

See ?save for one way to save models for later re-reading.

I don't fully understand your next question, many of the modeling functions
have options for saving the data (see ?lm), but can you give an example for your
A, B, and C example? Then we will have a better chance of telling you what is
happening and what options can change what is happening.

--
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.snow at imail.org
801.408.8111

> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of Sharma, Dhruv
> Sent: Monday, September 29, 2008 1:52 PM
> To: r-help at r-project.org
> Subject: [R] persistence of model in R to a file
>
>  Hi,
>     Is there a way to save R models (glm, lm , rpart etc) in a file
> that
> be read in later?
>
>     I noticed models take up space. by space them off and removing them
> from memory it seems that would be useful.
>
>     Also why do the models keep a copy of all columns in the original
> data set even those columns are not in the model.
>     E.g. if I build a model on columns A, B even thought column C
> exists
> in the dataframe when I run the model against test data with only
> columns A and B then I get an error saying column C is missing from the
> data.
>
>     Anyway it seems like the models would be more compact without
> having
> to include within them data from the original data set used to build
> them but that is an aside.
>
>     If models can be persisted and loaded selectively later that would
> neat.
>
>
> thanks
> Dhruv
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at 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.