uttam.phulwale@tcs.com
2005-Oct-06 07:13 UTC
[R] how to use tune.knn() for dataset with missing values
Hi Everybody, i again have the problem in using tune.knn(), its giving an error saying missing values are not allowed.... again here is the script for BreastCancer Data, library(e1071) library(mda) trdata<-data.frame(train,row.names=NULL) attach(trdata) xtr <- subset(trdata, select = -Class) ytr <- Class bestpara <-tune.knn(xtr,ytr, k = 1:25, tunecontrol = tune.control(sampling = "cross")) and here i got the mentioned error. can anybody help me in this regard... Thanks & Regards, Uttam Phulwale Tata Consultancy Services Limited Mailto: uttam.phulwale@tcs.com Website: http://www.tcs.com [[alternative HTML version deleted]]
David Meyer
2005-Oct-06 14:37 UTC
[R] how to use tune.knn() for dataset with missing values
Well, since knn() can't handle incomplete data as it says, you can choose to either omit incomplete observations (e.g., using na.omit()), or to impute the data if the conditions are met (missingness at random, ...); see, e.g., packages cat, mix, norm, and e1071 for that. HTH, David ------------------------ Hi Everybody, i again have the problem in using tune.knn(), its giving an error saying missing values are not allowed.... again here is the script for BreastCancer Data, library(e1071) library(mda) trdata<-data.frame(train,row.names=NULL) attach(trdata) xtr <- subset(trdata, select = -Class) ytr <- Class bestpara <-tune.knn(xtr,ytr, k = 1:25, tunecontrol tune.control(sampling = "cross")) and here i got the mentioned error. can anybody help me in this regard... Thanks & Regards, Uttam Phulwale Tata Consultancy Services Limited Mailto: uttam.phulwale at tcs.com Website: http://www.tcs.com