Chapter 68: Test from $train or $test ?

[gwierzchowski]

I'm not sure if this is on purpose or typo.
At very beginning of Chapter 68 there is code:

set.seed(123)
index <- sample(nrow(mnist$train$images), 10000)
x <- mnist$train$images[index,]
y <- factor(mnist$train$labels[index])

index <- sample(nrow(mnist$train$images), 1000)
x_test <- mnist$train$images[index,]
y_test <- factor(mnist$train$labels[index])

I'm not sure if instructions in second group maybe should use mnist$test?

[dedelman04]

Additionally, when fitting with the entire model, the text says to use k = 5, when k = 3 is shown to have the most accuracy based on the training:

fit_knn<- knn3(x[ ,col_index], y, k = 5)

(also a comma in y[index, ] in the n, b code sample to test out the model training should be outside the brackets)

[gillsignals]

Both issues have been addressed. @dedelman04 - next time it would be best if you open a new issue for a separate problem. Thanks!