Update YOLOv3 and YOLOv5 YAMLs (#7574)

Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com>

ultralytics/nn/modules/head.py

@@ -11,7 +11,7 @@ from ultralytics.utils.tal import TORCH_1_10, dist2bbox, dist2rbox, make_anchors
 from .block import DFL, Proto
 from .conv import Conv
 from .transformer import MLP, DeformableTransformerDecoder, DeformableTransformerDecoderLayer
-from .utils import bias_init_with_prob, linear_init_
+from .utils import bias_init_with_prob, linear_init
 
 __all__ = "Detect", "Segment", "Pose", "Classify", "OBB", "RTDETRDecoder"
 
@@ -417,18 +417,18 @@ class RTDETRDecoder(nn.Module):
         """Initializes or resets the parameters of the model's various components with predefined weights and biases."""
         # Class and bbox head init
         bias_cls = bias_init_with_prob(0.01) / 80 * self.nc
-        # NOTE: the weight initialization in `linear_init_` would cause NaN when training with custom datasets.
-        # linear_init_(self.enc_score_head)
+        # NOTE: the weight initialization in `linear_init` would cause NaN when training with custom datasets.
+        # linear_init(self.enc_score_head)
         constant_(self.enc_score_head.bias, bias_cls)
         constant_(self.enc_bbox_head.layers[-1].weight, 0.0)
         constant_(self.enc_bbox_head.layers[-1].bias, 0.0)
         for cls_, reg_ in zip(self.dec_score_head, self.dec_bbox_head):
-            # linear_init_(cls_)
+            # linear_init(cls_)
             constant_(cls_.bias, bias_cls)
             constant_(reg_.layers[-1].weight, 0.0)
             constant_(reg_.layers[-1].bias, 0.0)
 
-        linear_init_(self.enc_output[0])
+        linear_init(self.enc_output[0])
         xavier_uniform_(self.enc_output[0].weight)
         if self.learnt_init_query:
             xavier_uniform_(self.tgt_embed.weight)

ultralytics/nn/modules/utils.py

@@ -23,7 +23,7 @@ def bias_init_with_prob(prior_prob=0.01):
     return float(-np.log((1 - prior_prob) / prior_prob))  # return bias_init
 
 
-def linear_init_(module):
+def linear_init(module):
     """Initialize the weights and biases of a linear module."""
     bound = 1 / math.sqrt(module.weight.shape[0])
     uniform_(module.weight, -bound, bound)