Updated inceptionV3 to accept different sized images (Adaptive avg pool) (#744)

* Updated inceptionV3 to accept different sized images (Adaptive avg pool)

The update allows inceptionV3 to process images larger or smaller than prescribed image size (299x299) using adaptive average pooling. Will be useful while finetuning or testing on different resolution images.

* Update inception.py

Author: ekagra-ranjan

torchvision/models/inception.py

@@ -118,7 +118,8 @@ def forward(self, x):
         # 8 x 8 x 2048
         x = self.Mixed_7c(x)
         # 8 x 8 x 2048
-        x = F.avg_pool2d(x, kernel_size=8)
+        # Adaptive average pooling
+        x = F.adaptive_avg_pool2d(x, (1, 1))
         # 1 x 1 x 2048
         x = F.dropout(x, training=self.training)
         # 1 x 1 x 2048
@@ -311,6 +312,9 @@ def forward(self, x):
         # 5 x 5 x 128
         x = self.conv1(x)
         # 1 x 1 x 768
+        # Adaptive average pooling
+        x = F.adaptive_avg_pool2d(x, (1, 1))
+        # 1 x 1 x 768
         x = x.view(x.size(0), -1)
         # 768
         x = self.fc(x)