model_output = tf.subtract(tf.matmul(x_data, A), b) should be tf.add
MingleiLI opened this issue · comments
In linear SVM implementation of chapter 4,
model_output = tf.subtract(tf.matmul(x_data, A), b)
should be tf.add(tf.matmul(x_data, A), b)
Thanks @MingleiLI !
Technically the add/subtract issue is invariant under a b* = -b transformation. But since wikipedia has it the opposite way, and you're not the first to raise this issue, I'm going to align it with wikipedia so that the issues will stop.
I'm going to close this and update the linear SVM this week. Check back in about a week and see if it's updated - if not, feel free to reopen.