mirror of
https://github.com/ultralytics/yolov5.git
synced 2025-06-03 14:49:29 +08:00
34cf749958
* Update LICENSE to AGPL-3.0 This pull request updates the license of the YOLOv5 project from GNU General Public License v3.0 (GPL-3.0) to GNU Affero General Public License v3.0 (AGPL-3.0). We at Ultralytics have decided to make this change in order to better protect our intellectual property and ensure that any modifications made to the YOLOv5 source code will be shared back with the community when used over a network. AGPL-3.0 is very similar to GPL-3.0, but with an additional clause to address the use of software over a network. This change ensures that if someone modifies YOLOv5 and provides it as a service over a network (e.g., through a web application or API), they must also make the source code of their modified version available to users of the service. This update includes the following changes: - Replace the `LICENSE` file with the AGPL-3.0 license text - Update the license reference in the `README.md` file - Update the license headers in source code files We believe that this change will promote a more collaborative environment and help drive further innovation within the YOLOv5 community. Please review the changes and let us know if you have any questions or concerns. Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com> * Update headers to AGPL-3.0 --------- Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com>
104 lines
3.4 KiB
Python
104 lines
3.4 KiB
Python
# YOLOv5 🚀 by Ultralytics, AGPL-3.0 license
|
|
"""
|
|
Activation functions
|
|
"""
|
|
|
|
import torch
|
|
import torch.nn as nn
|
|
import torch.nn.functional as F
|
|
|
|
|
|
class SiLU(nn.Module):
|
|
# SiLU activation https://arxiv.org/pdf/1606.08415.pdf
|
|
@staticmethod
|
|
def forward(x):
|
|
return x * torch.sigmoid(x)
|
|
|
|
|
|
class Hardswish(nn.Module):
|
|
# Hard-SiLU activation
|
|
@staticmethod
|
|
def forward(x):
|
|
# return x * F.hardsigmoid(x) # for TorchScript and CoreML
|
|
return x * F.hardtanh(x + 3, 0.0, 6.0) / 6.0 # for TorchScript, CoreML and ONNX
|
|
|
|
|
|
class Mish(nn.Module):
|
|
# Mish activation https://github.com/digantamisra98/Mish
|
|
@staticmethod
|
|
def forward(x):
|
|
return x * F.softplus(x).tanh()
|
|
|
|
|
|
class MemoryEfficientMish(nn.Module):
|
|
# Mish activation memory-efficient
|
|
class F(torch.autograd.Function):
|
|
|
|
@staticmethod
|
|
def forward(ctx, x):
|
|
ctx.save_for_backward(x)
|
|
return x.mul(torch.tanh(F.softplus(x))) # x * tanh(ln(1 + exp(x)))
|
|
|
|
@staticmethod
|
|
def backward(ctx, grad_output):
|
|
x = ctx.saved_tensors[0]
|
|
sx = torch.sigmoid(x)
|
|
fx = F.softplus(x).tanh()
|
|
return grad_output * (fx + x * sx * (1 - fx * fx))
|
|
|
|
def forward(self, x):
|
|
return self.F.apply(x)
|
|
|
|
|
|
class FReLU(nn.Module):
|
|
# FReLU activation https://arxiv.org/abs/2007.11824
|
|
def __init__(self, c1, k=3): # ch_in, kernel
|
|
super().__init__()
|
|
self.conv = nn.Conv2d(c1, c1, k, 1, 1, groups=c1, bias=False)
|
|
self.bn = nn.BatchNorm2d(c1)
|
|
|
|
def forward(self, x):
|
|
return torch.max(x, self.bn(self.conv(x)))
|
|
|
|
|
|
class AconC(nn.Module):
|
|
r""" ACON activation (activate or not)
|
|
AconC: (p1*x-p2*x) * sigmoid(beta*(p1*x-p2*x)) + p2*x, beta is a learnable parameter
|
|
according to "Activate or Not: Learning Customized Activation" <https://arxiv.org/pdf/2009.04759.pdf>.
|
|
"""
|
|
|
|
def __init__(self, c1):
|
|
super().__init__()
|
|
self.p1 = nn.Parameter(torch.randn(1, c1, 1, 1))
|
|
self.p2 = nn.Parameter(torch.randn(1, c1, 1, 1))
|
|
self.beta = nn.Parameter(torch.ones(1, c1, 1, 1))
|
|
|
|
def forward(self, x):
|
|
dpx = (self.p1 - self.p2) * x
|
|
return dpx * torch.sigmoid(self.beta * dpx) + self.p2 * x
|
|
|
|
|
|
class MetaAconC(nn.Module):
|
|
r""" ACON activation (activate or not)
|
|
MetaAconC: (p1*x-p2*x) * sigmoid(beta*(p1*x-p2*x)) + p2*x, beta is generated by a small network
|
|
according to "Activate or Not: Learning Customized Activation" <https://arxiv.org/pdf/2009.04759.pdf>.
|
|
"""
|
|
|
|
def __init__(self, c1, k=1, s=1, r=16): # ch_in, kernel, stride, r
|
|
super().__init__()
|
|
c2 = max(r, c1 // r)
|
|
self.p1 = nn.Parameter(torch.randn(1, c1, 1, 1))
|
|
self.p2 = nn.Parameter(torch.randn(1, c1, 1, 1))
|
|
self.fc1 = nn.Conv2d(c1, c2, k, s, bias=True)
|
|
self.fc2 = nn.Conv2d(c2, c1, k, s, bias=True)
|
|
# self.bn1 = nn.BatchNorm2d(c2)
|
|
# self.bn2 = nn.BatchNorm2d(c1)
|
|
|
|
def forward(self, x):
|
|
y = x.mean(dim=2, keepdims=True).mean(dim=3, keepdims=True)
|
|
# batch-size 1 bug/instabilities https://github.com/ultralytics/yolov5/issues/2891
|
|
# beta = torch.sigmoid(self.bn2(self.fc2(self.bn1(self.fc1(y))))) # bug/unstable
|
|
beta = torch.sigmoid(self.fc2(self.fc1(y))) # bug patch BN layers removed
|
|
dpx = (self.p1 - self.p2) * x
|
|
return dpx * torch.sigmoid(beta * dpx) + self.p2 * x
|