Update face_recognition_model.py

app/Hackathon_setup/face_recognition.py

@@ -1,3 +1,4 @@
+from . import face_recognition_model
 import numpy as np
 import cv2
 from matplotlib import pyplot as plt
@@ -71,21 +72,30 @@ def get_similarity(img1, img2):
     ##the same path as this file, we recommend to put in the same directory                 ##
     ##########################################################################################
     ##########################################################################################
-    feature_net = Siamese()
-    model = torch.load(current_path + '/siamese_model.t7', map_location=device)
-    feature_net.load_state_dict(model['net_dict'])
-
-   
+ 
 
-    output1, output2 = feature_net((face1).to(device), (face2).to(device))
-    euclidean_distance = F.pairwise_distance(output1, output2)
+    feature_net = Siamese()
+    ckpt = torch.load(current_path + '/siamese_model.t7', map_location=device)
+    # model_path = current_path + "/Hackathon-setup/siamese_model.t7"
+    feature_net.load_state_dict(ckpt['net_dict'])
+    # model.eval()
 
-  
-    return euclidean_distance.item()
-    # YOUR CODE HERE, load the model
+    with torch.no_grad():
+        output1, output2 = feature_net(face1.to(device), face2.to(device))
+        # Calculate similarity measure - for instance, using cosine similarity
+        euclidean_distance = F.pairwise_distance(output1, output2)
     
-    # YOUR CODE HERE, return similarity measure using your model
+    return euclidean_distance.item()
+
+    # ckpt = torch.load(current_path + "/Hackathon-setup/siamese_model.t7", map_location=device)
+    # # YOUR CODE HERE, load the model
+    # similarity_measure = ckpt(face1.to(device), face2.to(device))
+    # # YOUR CODE HERE, return similarity measure using your model
     
+    # return similarity_measure
+
+
+    # Load the Siamese network model
     
     
 #1) Image captured from mobile is passed as parameter to this function in the API call, It returns the face class in the string form ex: "Person1"

app/Hackathon_setup/face_recognition_model.py

@@ -14,10 +14,9 @@ trnscm = transforms.Compose([transforms.Resize((100,100)), transforms.ToTensor()
 class Siamese(torch.nn.Module):
     def __init__(self):
         super(Siamese, self).__init__()
-        # super(SiameseNetwork, self).__init__()
+        #YOUR CODE HERE
         self.cnn1 = nn.Sequential(
-            nn.ReflectionPad2d(1),
-            # Pads the input tensor using the reflection of the input boundary, it similar to the padding.
+            nn.ReflectionPad2d(1),       #Pads the input tensor using the reflection of the input boundary, it similar to the padding.
             nn.Conv2d(1, 4, kernel_size=3),
             nn.ReLU(inplace=True),
             nn.BatchNorm2d(4),
@@ -27,6 +26,7 @@ class Siamese(torch.nn.Module):
             nn.ReLU(inplace=True),
             nn.BatchNorm2d(8),
 
+
             nn.ReflectionPad2d(1),
             nn.Conv2d(8, 8, kernel_size=3),
             nn.ReLU(inplace=True),
@@ -34,28 +34,26 @@ class Siamese(torch.nn.Module):
         )
 
         self.fc1 = nn.Sequential(
-            nn.Linear(8 * 100 * 100, 500),
+            nn.Linear(8*100*100, 500),
             nn.ReLU(inplace=True),
 
             nn.Linear(500, 500),
             nn.ReLU(inplace=True),
 
             nn.Linear(500, 5))
-        #YOUR CODE HERE
         
-        # forward_once is for one image. This can be used while classifying the face images
     def forward_once(self, x):
         output = self.cnn1(x)
         output = output.view(output.size()[0], -1)
         output = self.fc1(output)
         return output
 
-    def forward(self, x):
+    def forward(self, input1, input2):
         output1 = self.forward_once(input1)
         output2 = self.forward_once(input2)
         return output1, output2
-        # remove 'pass' once you have written your code
-        #YOUR CODE HERE
+
+
         
 ##########################################################################################################
 ## Sample classification network (Specify if you are using a pytorch classifier during the training)    ##
@@ -63,6 +61,14 @@ class Siamese(torch.nn.Module):
 ##########################################################################################################
 
 # YOUR CODE HERE for pytorch classifier
+num_of_classes = 6
+classifier = nn.Sequential(nn.Linear(256, 64),
+                           nn.BatchNorm1d(64),
+                           nn.ReLU(),
+                           nn.Linear(64, 32),
+                           nn.BatchNorm1d(32),
+                           nn.ReLU(),
+                           nn.Linear(32, num_of_classes))
 
 # Definition of classes as dictionary
-classes = ['person1','person2','person3','person4','person5','person6','person7']
+classes = ['person1','person2','person3','person4','person5','person6']