CODEWITHSUNDEEP
machine-learninglibsvm
Xiaobo Gu
Xiaobo Gu

Reputation: 189

What does y scaling mean in svm-scale

Does it mean to scale the value of the target(class),which is the first column of every instance?

Regards,

amber@cent64 libsvm-3.17]$ svm-scale
Usage: svm-scale [options] data_filename
options:
-l lower : x scaling lower limit (default -1)
-u upper : x scaling upper limit (default +1)
-y y_lower y_upper : y scaling limits (default: no y scaling)
-s save_filename : save scaling parameters to save_filename
-r restore_filename : restore scaling parameters from restore_filename

Upvotes: 1

Views: 1252

Answers (1)

jabaldonedo
jabaldonedo

Reputation: 26582

Yes it means to scale the output (target) value of each instance of your dataset in the following dataset:

y0 1:x00 2:x01 3:x02 ... n: x0n-1
y1 1:x10 2:x11 3:x12 ... n: x1n-1
...
yn 1:xn0 2:xn1 3:xn2 ... n: xnn-1

If you inspect the file svm-scale.c you will find that the formula that scales y data is:

value = y_lower + (y_upper-y_lower) * (value - y_min)/(y_max-y_min);

Upvotes: 2

Related Questions