`.detach()` on bias init (#8044)

make init bias better, 
I think `detach` is more safer than `data`

models/yolo.py

@@ -207,9 +207,9 @@ class Model(nn.Module):
         # cf = torch.bincount(torch.tensor(np.concatenate(dataset.labels, 0)[:, 0]).long(), minlength=nc) + 1.
         m = self.model[-1]  # Detect() module
         for mi, s in zip(m.m, m.stride):  # from
-            b = mi.bias.view(m.na, -1)  # conv.bias(255) to (3,85)
-            b.data[:, 4] += math.log(8 / (640 / s) ** 2)  # obj (8 objects per 640 image)
-            b.data[:, 5:] += math.log(0.6 / (m.nc - 0.999999)) if cf is None else torch.log(cf / cf.sum())  # cls
+            b = mi.bias.view(m.na, -1).detach()  # conv.bias(255) to (3,85)
+            b[:, 4] += math.log(8 / (640 / s) ** 2)  # obj (8 objects per 640 image)
+            b[:, 5:] += math.log(0.6 / (m.nc - 0.999999)) if cf is None else torch.log(cf / cf.sum())  # cls
             mi.bias = torch.nn.Parameter(b.view(-1), requires_grad=True)
 
     def _print_biases(self):