scottyli·vor 9 Jahren·discussWhat are the downsides of transfer learning? How can it fail?And do you just arbitrarily select the "cut off output layer" for the pretrained model when retraining with your own data on new layers?
scottyli·vor 11 Jahren·discusspg's essay on "How to disagree" (http://www.paulgraham.com/disagree.html) also serves as a useful guideline.grep 'u r a fag' | rm;
scottyli·vor 13 Jahren·discussOriginal quote from Arnold Rothstein (Boardwalk Empire)http://youtu.be/6WYsM1nIbKs?t=52s
And do you just arbitrarily select the "cut off output layer" for the pretrained model when retraining with your own data on new layers?