We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
I wanna write transformer module ,some subclass need rewrite like python
//class PositionalEncoding(nnModule //): // def __init__(self, d_model, max_len = 28 * 28): // super // (PositionalEncoding, self).__init__() // self.encoding = torch.zeros(max_len, d_model) // position = torch.arange(0, max_len, dtype = torch.float).unsqueeze(1) // div_term = torch.exp(torch.arange(0, d_model, 2).float() * (-torch.log(torch.tensor(10000.0)) / d_model)) // self.encoding[: // , 0 :: 2 // ] = torch.sin(position * div_term) // self.encoding[: // , 1 :: 2 // ] = torch.cos(position * div_term) // self.encoding = self.encoding.unsqueeze(0) // // def forward(self, x): // return x + self.encoding[: // ,: x.size(1) // ].to(x.device)
but in scala ,torch.zeros() dtype how to pass correct dtype on Model class init not in apply(),and tensor slice select
class PositionalEncoding[D <: BFloat16 | Float32 : Default](d_model: Long, max_len: Long = 28 * 28) extends HasParams[D] { val arr = Seq(max_len,d_model) var encoding = torch.zeros(size = arr.map(_.toInt), dtype = D.default) val position = torch.arange(0, max_len, dtype = torch.float).unsqueeze(1) val div_term = torch.exp(torch.arange(0, d_model, 2).float() * (-torch.log(torch.tensor(10000.0)) / d_model)) // encoding[::, 0 :: 2] = torch.sin(position * div_term) // encoding[::, 1 :: 2] = torch.cos(position * div_term) encoding[::,torch.indexing.::(0,2)] )= torch.sin(position * div_term) encoding[::, 1 :: 2] = torch.cos(position * div_term) encoding = encoding.unsqueeze(0) def apply(x: Tensor[D]): Tensor[D] = x + encoding[::,:: ].to(x.device) }
thanks for your reply
The text was updated successfully, but these errors were encountered:
No branches or pull requests
I wanna write transformer module ,some subclass need rewrite
like python
but in scala ,torch.zeros() dtype how to pass correct dtype on Model class init not in apply(),and tensor slice select
thanks for your reply
The text was updated successfully, but these errors were encountered: