Update model.py

model.py

@@ -1,21 +1,14 @@
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
-import torchvision
 
-from resnet import Resnet18
-# from modules.bn import InPlaceABNSync as BatchNorm2d
+from resnet import Resnet18  # Ensure that the Resnet18 class is correctly defined in this module
 
 
 class ConvBNReLU(nn.Module):
-    def __init__(self, in_chan, out_chan, ks=3, stride=1, padding=1, *args, **kwargs):
+    def __init__(self, in_chan, out_chan, ks=3, stride=1, padding=1):
         super(ConvBNReLU, self).__init__()
-        self.conv = nn.Conv2d(in_chan,
-                out_chan,
-                kernel_size = ks,
-                stride = stride,
-                padding = padding,
-                bias = False)
+        self.conv = nn.Conv2d(in_chan, out_chan, kernel_size=ks, stride=stride, padding=padding, bias=False)
         self.bn = nn.BatchNorm2d(out_chan)
         self.init_weight()
 
@@ -25,13 +18,13 @@ class ConvBNReLU(nn.Module):
         return x
 
     def init_weight(self):
-        for ly in self.children():
-            if isinstance(ly, nn.Conv2d):
-                nn.init.kaiming_normal_(ly.weight, a=1)
-                if not ly.bias is None: nn.init.constant_(ly.bias, 0)
+        nn.init.kaiming_normal_(self.conv.weight, a=1)
+        if self.conv.bias is not None:
+            nn.init.constant_(self.conv.bias, 0)
+
 
 class BiSeNetOutput(nn.Module):
-    def __init__(self, in_chan, mid_chan, n_classes, *args, **kwargs):
+    def __init__(self, in_chan, mid_chan, n_classes):
         super(BiSeNetOutput, self).__init__()
         self.conv = ConvBNReLU(in_chan, mid_chan, ks=3, stride=1, padding=1)
         self.conv_out = nn.Conv2d(mid_chan, n_classes, kernel_size=1, bias=False)
@@ -43,28 +36,23 @@ class BiSeNetOutput(nn.Module):
         return x
 
     def init_weight(self):
-        for ly in self.children():
-            if isinstance(ly, nn.Conv2d):
-                nn.init.kaiming_normal_(ly.weight, a=1)
-                if not ly.bias is None: nn.init.constant_(ly.bias, 0)
+        nn.init.kaiming_normal_(self.conv_out.weight, a=1)
+        if self.conv_out.bias is not None:
+            nn.init.constant_(self.conv_out.bias, 0)
 
     def get_params(self):
-        wd_params, nowd_params = [], []
-        for name, module in self.named_modules():
-            if isinstance(module, nn.Linear) or isinstance(module, nn.Conv2d):
-                wd_params.append(module.weight)
-                if not module.bias is None:
-                    nowd_params.append(module.bias)
-            elif isinstance(module, nn.BatchNorm2d):
-                nowd_params += list(module.parameters())
+        wd_params = [self.conv_out.weight]
+        nowd_params = []
+        if self.conv_out.bias is not None:
+            nowd_params.append(self.conv_out.bias)
         return wd_params, nowd_params
 
 
 class AttentionRefinementModule(nn.Module):
-    def __init__(self, in_chan, out_chan, *args, **kwargs):
+    def __init__(self, in_chan, out_chan):
         super(AttentionRefinementModule, self).__init__()
         self.conv = ConvBNReLU(in_chan, out_chan, ks=3, stride=1, padding=1)
-        self.conv_atten = nn.Conv2d(out_chan, out_chan, kernel_size= 1, bias=False)
+        self.conv_atten = nn.Conv2d(out_chan, out_chan, kernel_size=1, bias=False)
         self.bn_atten = nn.BatchNorm2d(out_chan)
         self.sigmoid_atten = nn.Sigmoid()
         self.init_weight()
@@ -79,14 +67,13 @@ class AttentionRefinementModule(nn.Module):
         return out
 
     def init_weight(self):
-        for ly in self.children():
-            if isinstance(ly, nn.Conv2d):
-                nn.init.kaiming_normal_(ly.weight, a=1)
-                if not ly.bias is None: nn.init.constant_(ly.bias, 0)
+        nn.init.kaiming_normal_(self.conv_atten.weight, a=1)
+        if self.conv_atten.bias is not None:
+            nn.init.constant_(self.conv_atten.bias, 0)
 
 
 class ContextPath(nn.Module):
-    def __init__(self, *args, **kwargs):
+    def __init__(self):
         super(ContextPath, self).__init__()
         self.resnet = Resnet18()
         self.arm16 = AttentionRefinementModule(256, 128)
@@ -95,8 +82,6 @@ class ContextPath(nn.Module):
         self.conv_head16 = ConvBNReLU(128, 128, ks=3, stride=1, padding=1)
         self.conv_avg = ConvBNReLU(512, 128, ks=1, stride=1, padding=0)
 
-        self.init_weight()
-
     def forward(self, x):
         H0, W0 = x.size()[2:]
         feat8, feat16, feat32 = self.resnet(x)
@@ -118,77 +103,22 @@ class ContextPath(nn.Module):
         feat16_up = F.interpolate(feat16_sum, (H8, W8), mode='nearest')
         feat16_up = self.conv_head16(feat16_up)
 
-        return feat8, feat16_up, feat32_up  # x8, x8, x16
-
-    def init_weight(self):
-        for ly in self.children():
-            if isinstance(ly, nn.Conv2d):
-                nn.init.kaiming_normal_(ly.weight, a=1)
-                if not ly.bias is None: nn.init.constant_(ly.bias, 0)
-
-    def get_params(self):
-        wd_params, nowd_params = [], []
-        for name, module in self.named_modules():
-            if isinstance(module, (nn.Linear, nn.Conv2d)):
-                wd_params.append(module.weight)
-                if not module.bias is None:
-                    nowd_params.append(module.bias)
-            elif isinstance(module, nn.BatchNorm2d):
-                nowd_params += list(module.parameters())
-        return wd_params, nowd_params
-
-
-### This is not used, since I replace this with the resnet feature with the same size
-class SpatialPath(nn.Module):
-    def __init__(self, *args, **kwargs):
-        super(SpatialPath, self).__init__()
-        self.conv1 = ConvBNReLU(3, 64, ks=7, stride=2, padding=3)
-        self.conv2 = ConvBNReLU(64, 64, ks=3, stride=2, padding=1)
-        self.conv3 = ConvBNReLU(64, 64, ks=3, stride=2, padding=1)
-        self.conv_out = ConvBNReLU(64, 128, ks=1, stride=1, padding=0)
-        self.init_weight()
-
-    def forward(self, x):
-        feat = self.conv1(x)
-        feat = self.conv2(feat)
-        feat = self.conv3(feat)
-        feat = self.conv_out(feat)
-        return feat
+        return feat8, feat16_up, feat32_up
 
     def init_weight(self):
         for ly in self.children():
             if isinstance(ly, nn.Conv2d):
                 nn.init.kaiming_normal_(ly.weight, a=1)
-                if not ly.bias is None: nn.init.constant_(ly.bias, 0)
-
-    def get_params(self):
-        wd_params, nowd_params = [], []
-        for name, module in self.named_modules():
-            if isinstance(module, nn.Linear) or isinstance(module, nn.Conv2d):
-                wd_params.append(module.weight)
-                if not module.bias is None:
-                    nowd_params.append(module.bias)
-            elif isinstance(module, nn.BatchNorm2d):
-                nowd_params += list(module.parameters())
-        return wd_params, nowd_params
+                if ly.bias is not None:
+                    nn.init.constant_(ly.bias, 0)
 
 
 class FeatureFusionModule(nn.Module):
-    def __init__(self, in_chan, out_chan, *args, **kwargs):
+    def __init__(self, in_chan, out_chan):
         super(FeatureFusionModule, self).__init__()
         self.convblk = ConvBNReLU(in_chan, out_chan, ks=1, stride=1, padding=0)
-        self.conv1 = nn.Conv2d(out_chan,
-                out_chan//4,
-                kernel_size = 1,
-                stride = 1,
-                padding = 0,
-                bias = False)
-        self.conv2 = nn.Conv2d(out_chan//4,
-                out_chan,
-                kernel_size = 1,
-                stride = 1,
-                padding = 0,
-                bias = False)
+        self.conv1 = nn.Conv2d(out_chan, out_chan // 4, kernel_size=1, stride=1, padding=0, bias=False)
+        self.conv2 = nn.Conv2d(out_chan // 4, out_chan, kernel_size=1, stride=1, padding=0, bias=False)
         self.relu = nn.ReLU(inplace=True)
         self.sigmoid = nn.Sigmoid()
         self.init_weight()
@@ -206,38 +136,27 @@ class FeatureFusionModule(nn.Module):
         return feat_out
 
     def init_weight(self):
-        for ly in self.children():
-            if isinstance(ly, nn.Conv2d):
-                nn.init.kaiming_normal_(ly.weight, a=1)
-                if not ly.bias is None: nn.init.constant_(ly.bias, 0)
-
-    def get_params(self):
-        wd_params, nowd_params = [], []
-        for name, module in self.named_modules():
-            if isinstance(module, nn.Linear) or isinstance(module, nn.Conv2d):
-                wd_params.append(module.weight)
-                if not module.bias is None:
-                    nowd_params.append(module.bias)
-            elif isinstance(module, nn.BatchNorm2d):
-                nowd_params += list(module.parameters())
-        return wd_params, nowd_params
+        nn.init.kaiming_normal_(self.conv1.weight, a=1)
+        if self.conv1.bias is not None:
+            nn.init.constant_(self.conv1.bias, 0)
+        nn.init.kaiming_normal_(self.conv2.weight, a=1)
+        if self.conv2.bias is not None:
+            nn.init.constant_(self.conv2.bias, 0)
 
 
 class BiSeNet(nn.Module):
-    def __init__(self, n_classes, *args, **kwargs):
+    def __init__(self, n_classes):
         super(BiSeNet, self).__init__()
         self.cp = ContextPath()
-        ## here self.sp is deleted
         self.ffm = FeatureFusionModule(256, 256)
         self.conv_out = BiSeNetOutput(256, 256, n_classes)
         self.conv_out16 = BiSeNetOutput(128, 64, n_classes)
         self.conv_out32 = BiSeNetOutput(128, 64, n_classes)
-        self.init_weight()
 
     def forward(self, x):
         H, W = x.size()[2:]
-        feat_res8, feat_cp8, feat_cp16 = self.cp(x)  # here return res3b1 feature
-        feat_sp = feat_res8  # use res3b1 feature to replace spatial path feature
+        feat_res8, feat_cp8, feat_cp16 = self.cp(x)
+        feat_sp = feat_res8  # Using res3b1 feature as spatial path feature
         feat_fuse = self.ffm(feat_sp, feat_cp8)
 
         feat_out = self.conv_out(feat_fuse)
@@ -253,28 +172,9 @@ class BiSeNet(nn.Module):
         for ly in self.children():
             if isinstance(ly, nn.Conv2d):
                 nn.init.kaiming_normal_(ly.weight, a=1)
-                if not ly.bias is None: nn.init.constant_(ly.bias, 0)
+                if ly.bias is not None:
+                    nn.init.constant_(ly.bias, 0)
 
     def get_params(self):
         wd_params, nowd_params, lr_mul_wd_params, lr_mul_nowd_params = [], [], [], []
-        for name, child in self.named_children():
-            child_wd_params, child_nowd_params = child.get_params()
-            if isinstance(child, FeatureFusionModule) or isinstance(child, BiSeNetOutput):
-                lr_mul_wd_params += child_wd_params
-                lr_mul_nowd_params += child_nowd_params
-            else:
-                wd_params += child_wd_params
-                nowd_params += child_nowd_params
-        return wd_params, nowd_params, lr_mul_wd_params, lr_mul_nowd_params
-
-
-if __name__ == "__main__":
-    net = BiSeNet(19)
-    #net.cuda()
-    net.eval()
-    in_ten = torch.randn(16, 3, 640, 480)
-    out, out16, out32 = net(in_ten)
-    print(out.shape)
-
-    net.get_params()
-
+        for name, child in