k-NN analysis for mlb dataset does not specify k

[tibbles-and-tribbles]

In regards to Section 1.8 "Example: The mlb Dataset", the code for building the k-NN model does not specify a value for k.

w <- mlb[, c(4:6)]
z <- qeKNN(w, "Weight")

Furthermore, I could not find a default value for k listed in the docs for the qeML package.

[matloff]

You're right! I'll fix that today. Remember, the latest version is in github.com/matloff/qeML.

One can check arguments using the R args() function. In this case,

args(qeKNN)

function (data, yName, k = 25, scaleX = TRUE, smoothingFtn = mean,

yesYVal = NULL, expandVars = NULL, expandVals = NULL, holdout =

floor(min(1000,

    0.1 * nrow(data))))

So the default is 25. But yes, I need to fix the man page, thanks.

Norm

On Wed, Jan 24, 2024 at 11:59 AM Andrew Black @.***> wrote:

In regards to Section 1.8 "Example: The mlb Dataset", the code for building the k-NN model does not specify a value for k.

w <- mlb[, c(4:6)] z <- qeKNN(w, "Weight")

I could not find a default value for k listed in the docs for the qeML package.

— Reply to this email directly, view it on GitHub https://github.com/matloff/ArtOfML/issues/1, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABZ34ZJEYBGRB52CZW23473YQFRZVAVCNFSM6AAAAABCJK43XOVHI2DSMVQWIX3LMV43ASLTON2WKOZSGA4TQOJYG4ZDIMA . You are receiving this because you are subscribed to this thread.Message ID: @.***>