Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

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

Removed the redundant SiLUActivation class. #27136

Merged
merged 2 commits into from
Nov 2, 2023
Merged

Removed the redundant SiLUActivation class. #27136

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
4f09440
Removed the redundant SiLUActivation class and now use nn.functional.…
hisushanta Oct 30, 2023
1a5fa4f
I apologize for adding torch.functional.silu. I have replaced it with…
hisushanta Oct 30, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 2 additions & 15 deletions src/transformers/activations.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -137,19 +137,6 @@ def forward(self, input: Tensor) -> Tensor:
return 0.5 * input * (1 + torch.tanh(self.precomputed_constant * (input + 0.044715 * torch.pow(input, 3))))


class SiLUActivation(nn.Module):
"""
See Gaussian Error Linear Units (Hendrycks et al., https://arxiv.org/abs/1606.08415) where the SiLU (Sigmoid Linear
Unit) was originally introduced and coined, and see Sigmoid-Weighted Linear Units for Neural Network Function
Approximation in Reinforcement Learning (Elfwing et al., https://arxiv.org/abs/1702.03118) and Swish: a Self-Gated
Activation Function (Ramachandran et al., https://arxiv.org/abs/1710.05941v1) where the SiLU was experimented with
later.
"""

def forward(self, input: Tensor) -> Tensor:
return nn.functional.silu(input)


class MishActivation(nn.Module):
"""
See Mish: A Self-Regularized Non-Monotonic Activation Function (Misra., https://arxiv.org/abs/1908.08681). Also
Expand Down Expand Up @@ -226,8 +213,8 @@ def __getitem__(self, key):
"relu2": ReLUSquaredActivation,
"relu6": nn.ReLU6,
"sigmoid": nn.Sigmoid,
"silu": SiLUActivation,
"swish": SiLUActivation,
"silu": nn.SiLU,
"swish": nn.SiLU,
"tanh": nn.Tanh,
}
ACT2FN = ClassInstantier(ACT2CLS)
Expand Down