Closed ATheCoder closed 2 years ago
It seems like the xavier or the Kaiming which is used by PyTorch as a default is considered a good way to initialize the weights.
xavier
Kaiming
However, checking the code I realized you are changing that to a normal distribution.
Can I know the reason?
It follows the original implementation of MAML. I don't think that makes too big of a difference.
It seems like the
xavier
or theKaiming
which is used by PyTorch as a default is considered a good way to initialize the weights.However, checking the code I realized you are changing that to a normal distribution.
Can I know the reason?