MetaAconC(c2)_bias

models/common.py

@@ -47,6 +47,7 @@ def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True):  # ch_in, ch_out, k
         # self.act = Mish() if act is True else (act if isinstance(act, nn.Module) else nn.Identity())
         # self.act = AconC() if act is True else (act if isinstance(act, nn.Module) else nn.Identity())
         # self.act = MetaAconC() if act is True else (act if isinstance(act, nn.Module) else nn.Identity())
+        # self.act = SiLU_beta() if act is True else (act if isinstance(act, nn.Module) else nn.Identity())
         self.act = MetaAconC(c2) if act is True else (act if isinstance(act, nn.Module) else nn.Identity())
 
     def forward(self, x):

models/yolo.py

@@ -90,7 +90,7 @@ def __init__(self, cfg='yolov5s.yaml', ch=3, nc=None, anchors=None):  # model, i
         m = self.model[-1]  # Detect()
         if isinstance(m, Detect):
             s = 256  # 2x min stride
-            m.stride = torch.tensor([s / x.shape[-2] for x in self.forward(torch.zeros(1, ch, s, s))])  # forward
+            m.stride = torch.tensor([s / x.shape[-2] for x in self.forward(torch.zeros(2, ch, s, s))])  # forward
             m.anchors /= m.stride.view(-1, 1, 1)
             check_anchor_order(m)
             self.stride = m.stride
@@ -264,7 +264,7 @@ def parse_model(d, ch):  # model_dict, input_channels(3)
     # Create model
     model = Model(opt.cfg).to(device)
     model.train()
-                        
+
     # Profile
     # img = torch.rand(8 if torch.cuda.is_available() else 1, 3, 320, 320).to(device)
     # y = model(img, profile=True)

utils/activations.py

@@ -12,6 +12,15 @@ def forward(x):
         return x * torch.sigmoid(x)
 
 
+class SiLU_beta(nn.Module):  # SiLU() with B term: y = x * sigmoid(B * x)
+    def __init__(self, c1):
+        super().__init__()
+        self.beta = nn.Parameter(torch.ones(1, c1, 1, 1))
+
+    def forward(self, x):
+        return x * torch.sigmoid(self.beta * x)
+
+
 class Hardswish(nn.Module):  # export-friendly version of nn.Hardswish()
     @staticmethod
     def forward(x):
@@ -84,16 +93,14 @@ def __init__(self, c1, k=1, s=1, r=16):  # ch_in, kernel, stride, r
         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=False)
+        self.fc1 = nn.Conv2d(c1, c2, k, s, bias=True)
         self.bn1 = nn.BatchNorm2d(c2)
-        self.fc2 = nn.Conv2d(c2, c1, k, s, bias=False)
+        self.fc2 = nn.Conv2d(c2, c1, k, s, bias=True)
         self.bn2 = nn.BatchNorm2d(c1)
 
+    # batch-size 1 bug https://github.com/nmaac/acon/issues/4
     def forward(self, x):
         y = x.mean(dim=2, keepdims=True).mean(dim=3, keepdims=True)
-        if x.shape[0] > 1:  # batch-size 1 bug https://github.com/nmaac/acon/issues/4
-            beta = torch.sigmoid(self.bn2(self.fc2(self.bn1(self.fc1(y)))))
-        else:
-            beta = torch.sigmoid(self.fc2(self.fc1(y)))
+        beta = torch.sigmoid(self.bn2(self.fc2(self.bn1(self.fc1(y)))))
         dpx = (self.p1 - self.p2) * x
         return dpx * torch.sigmoid(beta * dpx) + self.p2 * x