scottyli
Re:How to use Deep Learning when you have Limited Data
on Thursday February 02, 2017 @08:59PM (#13556999)
What 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?
Re:New Hacker News Guideline: Avoid Gratuitous Negativity
on Friday April 03, 2015 @03:44PM (#9317963)
pg's essay on "How to disagree" (http://www.paulgraham.com/disagree.html) also serves as a useful guideline.
grep 'u r a fag' | rm;
Original quote from Arnold Rothstein (Boardwalk Empire)