huber loss equation in eager execution

[fsilavong]

Hi, I was going through the assignments and realised that the huber loss equation was multiplied by two here:

stanford-tensorflow-tutorials/examples/04_linreg_eager.py

Line 43 in 51e53da

return t ** 2 if tf.abs(t) <= m else m * (2 * tf.abs(t) - m)

Any reason for that?