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

Jump to bottom

[NFC] polish colossalai/kernel/jit/bias_gelu.py code style #946

Merged
merged 1 commit into from
May 13, 2022
Merged

[NFC] polish colossalai/kernel/jit/bias_gelu.py code style #946

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
12 changes: 8 additions & 4 deletions colossalai/kernel/jit/bias_gelu.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,5 @@
import torch


###### BIAS GELU FUSION/ NO AUTOGRAD ################
# 1/sqrt(2*pi)-> 0.3989423
# 1/sqrt(2) -> 0.70710678
Expand All @@ -9,10 +8,12 @@
# actual gelu is:
# x * 0.5 * (1.0 + torch.erf(x * 0.70710678))


@torch.jit.script
def bias_gelu(bias, y):
x = bias + y
return x * 0.5 * (1.0 + torch.tanh(0.79788456 * x * (1 + 0.044715 * x * x)))
return x * 0.5 * (1.0 + torch.tanh(0.79788456 * x * (1 + 0.044715 * x * x)))


# gradient of tanh approximation of gelu
# gradient of actual gelu is:
Expand All @@ -23,9 +24,11 @@ def bias_gelu_back(g, bias, y):
tanh_out = torch.tanh(0.79788456 * x * (1 + 0.044715 * x * x))
# sqrt(2/pi) * 3 * 0.044715 -> 0.1070322243
ff = 0.5 * x * ((1 - tanh_out * tanh_out) * (0.79788456 + 0.1070322243 * x * x)) + 0.5 * (1 + tanh_out)
return ff*g
return ff * g


class GeLUFunction(torch.autograd.Function):

@staticmethod
# bias is an optional argument
def forward(ctx, input, bias):
Expand All @@ -38,4 +41,5 @@ def backward(ctx, grad_output):
tmp = bias_gelu_back(grad_output, bias, input)
return tmp, tmp

bias_gelu_impl = GeLUFunction.apply

bias_gelu_impl = GeLUFunction.apply