From 4e2d24602d246231694ba1b4d3bf3bd01f027ea4 Mon Sep 17 00:00:00 2001
From: Glenn Jocher <glenn.jocher@ultralytics.com>
Date: Sat, 4 Jul 2020 23:07:26 -0700
Subject: [PATCH] update yolo.py

---
 models/yolo.py | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/models/yolo.py b/models/yolo.py
index 7cc86b2..66bdb7d 100644
--- a/models/yolo.py
+++ b/models/yolo.py
@@ -59,10 +59,14 @@ class Model(nn.Module):
 
         # Build strides, anchors
         m = self.model[-1]  # Detect()
-        m.stride = torch.tensor([128 / x.shape[-2] for x in self.forward(torch.zeros(1, ch, 128, 128))])  # forward
-        m.anchors /= m.stride.view(-1, 1, 1)
-        check_anchor_order(m)
-        self.stride = m.stride
+        if isinstance(m, Detect):
+            s = 128  # 2x min stride
+            m.stride = torch.tensor([s / x.shape[-2] for x in self.forward(torch.zeros(1, ch, s, s))])  # forward
+            m.anchors /= m.stride.view(-1, 1, 1)
+            check_anchor_order(m)
+            self.stride = m.stride
+            self._initialize_biases()  # only run once
+            # print('Strides: %s' % m.stride.tolist())
 
         # Init weights, biases
         torch_utils.initialize_weights(self)
@@ -146,7 +150,7 @@ class Model(nn.Module):
 
 
 def parse_model(md, ch):  # model_dict, input_channels(3)
-    print('\n%3s%15s%3s%10s  %-40s%-30s' % ('', 'from', 'n', 'params', 'module', 'arguments'))
+    print('\n%3s%18s%3s%10s  %-40s%-30s' % ('', 'from', 'n', 'params', 'module', 'arguments'))
     anchors, nc, gd, gw = md['anchors'], md['nc'], md['depth_multiple'], md['width_multiple']
     na = (len(anchors[0]) // 2)  # number of anchors
     no = na * (nc + 5)  # number of outputs = anchors * (classes + 5)
@@ -161,7 +165,7 @@ def parse_model(md, ch):  # model_dict, input_channels(3)
                 pass
 
         n = max(round(n * gd), 1) if n > 1 else n  # depth gain
-        if m in [nn.Conv2d, Conv, Bottleneck, SPP, DWConv, MixConv2d, Focus, BottleneckCSP, CrossConv]:
+        if m in [nn.Conv2d, Conv, Bottleneck, SPP, DWConv, MixConv2d, Focus, CrossConv, BottleneckCSP, C3]:
             c1, c2 = ch[f], args[0]
 
             # Normal
@@ -182,7 +186,7 @@ def parse_model(md, ch):  # model_dict, input_channels(3)
             #     c2 = make_divisible(c2, 8) if c2 != no else c2
 
             args = [c1, c2, *args[1:]]
-            if m is BottleneckCSP:
+            if m in [BottleneckCSP, C3]:
                 args.insert(2, n)
                 n = 1
         elif m is nn.BatchNorm2d:
@@ -198,7 +202,7 @@ def parse_model(md, ch):  # model_dict, input_channels(3)
         t = str(m)[8:-2].replace('__main__.', '')  # module type
         np = sum([x.numel() for x in m_.parameters()])  # number params
         m_.i, m_.f, m_.type, m_.np = i, f, t, np  # attach index, 'from' index, type, number params
-        print('%3s%15s%3s%10.0f  %-40s%-30s' % (i, f, n, np, t, args))  # print
+        print('%3s%18s%3s%10.0f  %-40s%-30s' % (i, f, n, np, t, args))  # print
         save.extend(x % i for x in ([f] if isinstance(f, int) else f) if x != -1)  # append to savelist
         layers.append(m_)
         ch.append(c2)