Maybe this is what you want:
you are right about the re-allocating the tables, but you can subset your
table into a new one:
selection <- which(device_Prob_Vector > 0.5)
# or via sample: selection <- sample(num_Devices)
training_Set <- measurements[selection]
validation_Set <- measurements[-selection]
good luck
Bart
natekupp wrote:
>
> Hey all,
>
> I'm doing some work with machine learning on R (I'm a fairly new user of
> R), and I have a question about generating new tables from existing
> tables. I'm currently using a table of measurements I read in from a CSV
> file to generate training and validation data set tables for future use in
> a machine learning algorithm using the code:
>
> #generate probabilities to divide up training / validation data sets
> randomly
> device_Prob_Vector <- runif(num_Devices)
>
> #NULL-initialize training and validation sets. This seems like a bit of a
> hack...
> training_Set <- measurements[0]
> validation_Set <- measurements[0]
>
> #divide up the training and validation data sets from measurements.
> for ( i in 1:num_Devices)
> {
> if ( device_Prob_Vector[i] > 0.5 )
> {
> training_Set <- rbind(training_Set, measurements[i,])
> }
> else
> {
> validation_Set <- rbind(validation_Set, measurements[i,])
> }
> }
>
> This code works correctly, but takes quite a long time to execute. I
> suspect this is because rbind() is dynamically resizing the tables as it
> adds new rows to each table of data. Is there a way to pre-allocate
> memory for each of the two tables, and then shrink them after the loop has
> completed? Thanks for the help.
>
> ~Nate
>
--
View this message in context:
http://www.nabble.com/Simple-table-generation-question-tf4056042.html#a11522530
Sent from the R help mailing list archive at Nabble.com.
______________________________________________
R-help@stat.math.ethz.ch 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.
