search

Rongjiehuang / FastDiff

PyTorch Implementation of FastDiff (IJCAI'22)
408 stars 64 forks source link

Optimizer parameters #10

Closed yahshibu closed 2 years ago

yahshibu commented 2 years ago

Thank you for making your code publicly available!

I have a question about the optimizer parameters. The paper says β1 = 0.9, β2 = 0.98, ε= 1e−9 were used (Section 5.1). On the other hand, this repository is using the default values of AdamW optimizer β1 = 0.9, β2 = 0.999, ε= 1e−8. https://github.com/Rongjiehuang/FastDiff/blob/f737e17e344923a7ec60507b289abd756f76000a/modules/FastDiff/task/FastDiff.py#L119-L121

Which setup is your recommendation?

yahshibu commented 2 years ago

I'm sorry. I've noticed that β1 = 0.9, β2 = 0.98, ε= 1e−9 stated in the paper are the values for FastDiff-TTS, not for FastDiff vocoder.