Hey.
Thanks for the great project and code.
I am running the code by myself and I just wonder does the "lr" in the train.sh actually control anything?
In the model/base_model.py, the learning rates for G and D are specified by lr_g and lr_d (line 21),
and it seems that lr_g and lr_d are default values that are different from the reported number in the paper.
I just wonder is my understanding correct?
Hey. Thanks for the great project and code. I am running the code by myself and I just wonder does the "lr" in the train.sh actually control anything? In the model/base_model.py, the learning rates for G and D are specified by lr_g and lr_d (line 21), and it seems that lr_g and lr_d are default values that are different from the reported number in the paper. I just wonder is my understanding correct?