paper and code implementation are different
Closed this issue · 2 comments
chenslcool commented
Thank your for nice job!
I notice that you use dot-product to measure simlarity between prediction in paper while use KL_div in code.
code:
NRC_SFDA/office-home/train_tar.py
Lines 315 to 317 in 1c26160
I wonder why do you do this?
Thanks!
Albert0147 commented
Hi, if you check pytorch document of kl_div, you will find they are indentical, without log here is just dot product.
chenslcool commented
...thanks 🤣