Create models/custom_resnet.py

models/custom_resnet.py

@@ -0,0 +1,85 @@
+from __future__ import print_function
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+import torch.optim as optim
+from torchvision import datasets, transforms
+from tqdm import tqdm
+
+class Net(nn.Module):
+    def __init__(self):
+        super(Net, self).__init__()
+
+
+        self.prep_layer = nn.Sequential(
+            nn.Conv2d(in_channels=3, out_channels=64, kernel_size=(3, 3), padding=1, stride=1, bias=False),
+            nn.BatchNorm2d(64),
+            nn.ReLU(),
+        ) # output_size = 64,32,32
+
+        self.layer1_x = nn.Sequential(
+            nn.Conv2d(in_channels=64, out_channels=128, kernel_size=(3, 3), padding=1, stride=1, bias=False),
+            nn.MaxPool2d(2, 2),
+            nn.BatchNorm2d(128),
+            nn.ReLU(),
+        ) # output_size = 128,16,16
+
+        self.layer1_r = nn.Sequential(
+            nn.Conv2d(in_channels=128, out_channels=128, kernel_size=(3, 3), padding=1, stride=1, bias=False),
+            nn.BatchNorm2d(128),
+            nn.ReLU(),
+            nn.Conv2d(in_channels=128, out_channels=128, kernel_size=(3, 3), padding=1, stride=1, bias=False),
+            nn.BatchNorm2d(128),
+            nn.ReLU(),
+        ) #output_size = 128,16,16
+
+        self.layer2 = nn.Sequential(
+            nn.Conv2d(in_channels=128, out_channels=256, kernel_size=(3, 3), padding=1, stride=1, bias=False),
+            nn.MaxPool2d(2, 2),
+            nn.BatchNorm2d(256),
+            nn.ReLU(),
+        ) # output_size = 256,8,8
+
+
+        self.layer3_x = nn.Sequential(
+            nn.Conv2d(in_channels=256, out_channels=512, kernel_size=(3, 3), padding=1, stride=1, bias=False),
+            nn.MaxPool2d(2, 2),
+            nn.BatchNorm2d(512),
+            nn.ReLU(),
+        ) # output_size = 512,4,4
+
+        self.layer3_r = nn.Sequential(
+            nn.Conv2d(in_channels=512, out_channels=512, kernel_size=(3, 3), padding=1, stride=1, bias=False),
+            nn.BatchNorm2d(512),
+            nn.ReLU(),
+            nn.Conv2d(in_channels=512, out_channels=512, kernel_size=(3, 3), padding=1, stride=1, bias=False),
+            nn.BatchNorm2d(512),
+            nn.ReLU(),
+        ) #output_size = 512,4,4
+
+        self.last_maxpool = nn.Sequential(
+           nn.MaxPool2d(4, 4), #512
+
+        )
+        self.last_fc =  nn.Sequential(
+            nn.Linear(512,10,bias=False)
+        )
+
+    def forward(self, x):
+        x = self.prep_layer(x)
+        x = self.layer1_x(x)
+        x_layer1_identity = x.clone()
+        x = self.layer1_r(x)
+        #x = F.relu(x + x_layer1_identity)
+        x = x + x_layer1_identity
+        x = self.layer2(x)
+        x = self.layer3_x(x)
+        x_layer3_identity = x.clone()
+        x = self.layer3_r(x)
+        #x = F.relu(x + x_layer3_identity)
+        x = x + x_layer3_identity
+        x = self.last_maxpool(x)
+        x = x.view(-1,512)
+        x = self.last_fc(x)
+        x = x.view(-1, 10)
+        return F.log_softmax(x, dim=-1)