Backward compatibility support for legacy models (#17010)

Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>

ultralytics/nn/modules/head.py

@@ -28,6 +28,7 @@ class Detect(nn.Module):
     shape = None
     anchors = torch.empty(0)  # init
     strides = torch.empty(0)  # init
+    legacy = False  # backward compatibility for v3/v5/v8/v9 models
 
     def __init__(self, nc=80, ch=()):
         """Initializes the YOLO detection layer with specified number of classes and channels."""
@@ -41,13 +42,17 @@ class Detect(nn.Module):
         self.cv2 = nn.ModuleList(
             nn.Sequential(Conv(x, c2, 3), Conv(c2, c2, 3), nn.Conv2d(c2, 4 * self.reg_max, 1)) for x in ch
         )
-        self.cv3 = nn.ModuleList(
-            nn.Sequential(
-                nn.Sequential(DWConv(x, x, 3), Conv(x, c3, 1)),
-                nn.Sequential(DWConv(c3, c3, 3), Conv(c3, c3, 1)),
-                nn.Conv2d(c3, self.nc, 1),
+        self.cv3 = (
+            nn.ModuleList(nn.Sequential(Conv(x, c3, 3), Conv(c3, c3, 3), nn.Conv2d(c3, self.nc, 1)) for x in ch)
+            if self.legacy
+            else nn.ModuleList(
+                nn.Sequential(
+                    nn.Sequential(DWConv(x, x, 3), Conv(x, c3, 1)),
+                    nn.Sequential(DWConv(c3, c3, 3), Conv(c3, c3, 1)),
+                    nn.Conv2d(c3, self.nc, 1),
+                )
+                for x in ch
             )
-            for x in ch
         )
         self.dfl = DFL(self.reg_max) if self.reg_max > 1 else nn.Identity()
 

ultralytics/nn/tasks.py

@@ -936,6 +936,7 @@ def parse_model(d, ch, verbose=True):  # model_dict, input_channels(3)
     import ast
 
     # Args
+    legacy = True  # backward compatibility for v3/v5/v8/v9 models
     max_channels = float("inf")
     nc, act, scales = (d.get(x) for x in ("nc", "activation", "scales"))
     depth, width, kpt_shape = (d.get(x, 1.0) for x in ("depth_multiple", "width_multiple", "kpt_shape"))
@@ -1027,8 +1028,10 @@ def parse_model(d, ch, verbose=True):  # model_dict, input_channels(3)
             }:
                 args.insert(2, n)  # number of repeats
                 n = 1
-            if m is C3k2 and scale in "mlx":  # for M/L/X sizes
-                args[3] = True
+            if m is C3k2:  # for M/L/X sizes
+                legacy = False
+                if scale in "mlx":
+                    args[3] = True
         elif m is AIFI:
             args = [ch[f], *args]
         elif m in {HGStem, HGBlock}:
@@ -1047,6 +1050,8 @@ def parse_model(d, ch, verbose=True):  # model_dict, input_channels(3)
             args.append([ch[x] for x in f])
             if m is Segment:
                 args[2] = make_divisible(min(args[2], max_channels) * width, 8)
+            if m in {Detect, Segment, Pose, OBB}:
+                m.legacy = legacy
         elif m is RTDETRDecoder:  # special case, channels arg must be passed in index 1
             args.insert(1, [ch[x] for x in f])
         elif m is CBLinear: