Grokbase Groups R r-devel April 2012
I am developing a custom regression package, which accepts a formula object
as way of setting up the model matrix and response variable from a data
frame. For large data sets, I expect that going through R memory might be
too slow, so I'm thinking about reading the data directly into C (e.g. from
a database) and then parsing the formula inside of C (by calling an R
function such as model.frame()). Is this possible? An alternative solution
is to read the large data set into R in chunks, create the model matrix for
each chunk and save the matrix chunk to disk, and then load and piece
together these chunks inside of C. But my solution seems sub-optimal to me.
Any suggestions? Thank you!

View this message in context:
Sent from the R devel mailing list archive at

Search Discussions

Related Discussions

Discussion Navigation
viewthread | post
posts ‹ prev | 1 of 1 | next ›
Discussion Overview
groupr-devel @
postedApr 24, '12 at 4:41p
activeApr 24, '12 at 4:41p

1 user in discussion

Alireza Mahani: 1 post



site design / logo © 2022 Grokbase