[R] R doesn't recognize R_HOME value

Upgraded to 2.15.1 from 2.15.0 this morning, on Windows 7. I'm setting R_HOME in Control Panel, and before the upgrade, 2.15.0 recognized the value with no problems and would use it to find rprofile.site, etc. Now, after upgrade, neither .0 nor .1 recognize it at all. From the cmd line, Windows ind

Re: [R] R doesn't recognize R_HOME value

??? Windows 7 is where I'm seeing the problem. The root problem: while I have R_HOME legitimately specified within Windows 7, and issuing 'set R_HOME' from the command line returns exactly the path I've specified, doing a Sys.getenv('R_HOME') from R returns a completely different path. I ask, "W

Re: [R] R doesn't recognize R_HOME value

>From the beginning I'd set R_HOME from within the Advanced area of Control Panel, and never actually set it from within a shell. I'll take the advice given, however, and not set it to any value at all. I did feel this was an R question since things were working as expected until upgrading, then no

Re: [R] R doesn't recognize R_HOME value

I wanted to wrap this up, since I feel it's been resolved. R_HOME never did need to over-ridden by me, was not being over-ridden in spite of my attempts, and was never a factor during the long period when 'everything worked'. In fact, my entire ordeal was caused by removing a comment from the rpr

[R] Analyzing Poor Performance Using naiveBayes()

My data is 50,000 instances of about 200 predictor values, and for all 50,000 examples I have the actual class labels (binary). The data is quite unbalanced with about 10% or less of the examples having a positive outcome and the remainder, of course, negative. Nothing suggests the data has any ord

Re: [R] Analyzing Poor Performance Using naiveBayes()

Per your suggestion I ran chi.squared() against my training data and to my delight, found just 50 parameters that were non-zero influencers. I built the model through several iterations and found n = 12 to be the optimum for the training data. However, results still no so good for the test data. H

Re: [R] Analyzing Poor Performance Using naiveBayes()

As some additional information, I re-ran the model across the range of n = 50 to 150 (n being the 'top n' parameters returned by chi.squared), and this time used a completed different subset of the data for both training and test. Nearly identical results, with the typical train AUC about 0.98 and