Peter:
Yes, this is a bug. The line
if ([EMAIL PROTECTED] <= nrow(x))
should actually read
if (([EMAIL PROTECTED] <= length([EMAIL PROTECTED])) && ([EMAIL PROTECTED] <
[EMAIL PROTECTED] + 1]))
--> fixed upstream.
Thanks
David
--------
Hello,
The write.matrix.csr() function of the e1071 package contains a bug.
Try the following:
library(e1071)
m <- 1 - diag(10)
sm <- as.matrix.csr(m)
write.matrix.csr(sm)
The resulting file (out.dat) contains only the two lines below:
2:1 3:1 4:1 5:1 6:1 7:1 8:1 9:1 10:1
1:1 3:1 4:1 5:1 6:1 7:1 8:1 9:1 10:1
This is obviously wrong as the matrix m has 90 non-zero entries.
The function is implemented as
write.matrix.csr <- function (x, file="out.dat", y=NULL) {
on.exit(sink())
if(!require("methods")) stop("Could not load package
'methods'.")
if (!is.null(y) & (length(y) != nrow(x)))
stop(paste("Length of y (=", length(y),
") does not match number of rows of x (=",
nrow(x), ")!", sep=""))
sink(file)
for (i in 1:nrow(x)) {
if (!is.null(y)) cat (y[i],"")
if ([EMAIL PROTECTED] <= nrow(x))
for (j in [EMAIL PROTECTED]:([EMAIL PROTECTED]
- 1))
cat([EMAIL PROTECTED], ":", [EMAIL PROTECTED], "
", sep="")
cat("\n")
}
}
The bug is caused by the line
if ([EMAIL PROTECTED] <= nrow(x))
that is completely meaningless. In the above example nrow(x) is 10
and [EMAIL PROTECTED] is the vector
1 10 19 28 37 46 55 64 73 82 91
Entries of [EMAIL PROTECTED] has nothing to do with the number of rows in the
matrix.
To fix the bug this line must be removed.
Best regards,
______________________________________________
R-help@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.
