diff --git a/CaffeLoader.py b/CaffeLoader.py
index d09ccdd..ad2e7f6 100644
--- a/CaffeLoader.py
+++ b/CaffeLoader.py
@@ -193,28 +193,31 @@ def buildSequential(channel_list, pooling):
 'P': ['pool1', 'pool2', 'pool3', 'pool4', 'pool5'],
 }
 
+name_dict = {
+'vgg19': ['vgg-19', 'vgg19', 'vgg_19',],
+'vgg16': ['vgg-16', 'vgg16', 'vgg_16', 'fcn32s', 'pruning', 'sod'],
+}
+
 
 def modelSelector(model_file, pooling):
-    vgg_list = ["fcn32s", "pruning", "sod", "vgg"]
-    if any(name in model_file for name in vgg_list):
-        if "pruning" in model_file:
-            print("VGG-16 Architecture Detected")
-            print("Using The Channel Pruning Model")
-            cnn, layerList = VGG_PRUNED(buildSequential(channel_list['VGG-16p'], pooling)), vgg16_dict
-        elif "fcn32s" in model_file:
+    if any(name in model_file for name in ['vgg'] + name_dict['vgg16'] + name_dict['vgg19']):
+        if any(name in model_file for name in name_dict['vgg16']):
             print("VGG-16 Architecture Detected")
-            print("Using the fcn32s-heavy-pascal Model")
-            cnn, layerList = VGG_FCN32S(buildSequential(channel_list['VGG-16'], pooling)), vgg16_dict
-        elif "sod" in model_file:
-            print("VGG-16 Architecture Detected")
-            print("Using The SOD Fintune Model")
-            cnn, layerList = VGG_SOD(buildSequential(channel_list['VGG-16'], pooling)), vgg16_dict
-        elif "19" in model_file:
+            if "pruning" in model_file:
+                print("Using The Channel Pruning Model")
+                cnn, layerList = VGG_PRUNED(buildSequential(channel_list['VGG-16p'], pooling)), vgg16_dict
+            elif "fcn32s" in model_file:
+                print("Using the fcn32s-heavy-pascal Model")
+                cnn, layerList = VGG_FCN32S(buildSequential(channel_list['VGG-16'], pooling)), vgg16_dict
+            elif "sod" in model_file:
+                print("Using The SOD Fintune Model")
+                cnn, layerList = VGG_SOD(buildSequential(channel_list['VGG-16'], pooling)), vgg16_dict
+            elif "16" in model_file:
+                cnn, layerList = VGG(buildSequential(channel_list['VGG-16'], pooling)), vgg16_dict
+        elif any(name in model_file for name in name_dict['vgg19']):
             print("VGG-19 Architecture Detected")
-            cnn, layerList = VGG(buildSequential(channel_list['VGG-19'], pooling)), vgg19_dict
-        elif "16" in model_file:
-            print("VGG-16 Architecture Detected")
-            cnn, layerList = VGG(buildSequential(channel_list['VGG-16'], pooling)), vgg16_dict
+            if "19" in model_file:
+                cnn, layerList = VGG(buildSequential(channel_list['VGG-19'], pooling)), vgg19_dict
         else:
             raise ValueError("VGG architecture not recognized.")
     elif "nin" in model_file:
@@ -251,4 +254,4 @@ def loadCaffemodel(model_file, pooling, use_gpu, disable_check):
 
     print_loadcaffe(cnn, layerList)
 
-    return cnn, layerList
+    return cnn, layerList
\ No newline at end of file
diff --git a/INSTALL.md b/INSTALL.md
index 7d7c3c2..dfdde55 100644
--- a/INSTALL.md
+++ b/INSTALL.md
@@ -197,7 +197,7 @@ Then using https://pytorch.org/, get the correct pip command, paste it into the
 
 
 ```
-pip3 install torch===1.3.0 torchvision===0.4.1 -f https://download.pytorch.org/whl/torch_stable.html
+pip3 install torch===1.3.1 torchvision===0.4.2 -f https://download.pytorch.org/whl/torch_stable.html
 ```
 
 
diff --git a/README.md b/README.md
index 35cbe79..44e8948 100644
--- a/README.md
+++ b/README.md
@@ -195,6 +195,7 @@ path or a full absolute path.
   when using ADAM you will probably need to play with other parameters to get good results, especially
   the style weight, content weight, and learning rate.
 * `-learning_rate`: Learning rate to use with the ADAM optimizer. Default is 1e1.
+* `-normalize_weights`: If this flag is present, style and content weights will be divided by the number of channels for each layer. Idea from [PytorchNeuralStyleTransfer](https://github.com/leongatys/PytorchNeuralStyleTransfer).
 
 **Output options**:
 * `-output_image`: Name of the output image. Default is `out.png`.