how to decide the value of λ

[silentsaber]

Since the d_loss = fake-real+gp, how to decide the value of LAMBDA to get a reasonable gradient penalty.
I mean, if fake=-10, real=10, is gp=10 appropriate？or the gp should be smaller(e.g. gp=5)?

[hangexuexi]

i have the same question about it, did you solve this problem? at the start of my train, the wass loss is very small, about 0.01, so how should i set the lambda?

[cjblocker]

The original paper suggests sticking with 10

Penalty coefficient All experiments in this paper use λ = 10, which we found to work well across a variety of architectures and datasets ranging from toy tasks to large ImageNet CNNs.