tried testing (unsuccessfully)

1 files changed, 7 insertions, 3 deletions

diff --git a/losses.py b/losses.py
index 16ca278e..297bd2f1 100644
--- a/losses.py
+++ b/losses.py
@@ -65,8 +65,8 @@ class YourModel(tf.keras.Model):
                    activation="relu", name="block5_conv3"),
             AveragePooling2D(2, name="block5_pool"),
         ]
-       #  for layer in self.vgg16:
-       #         layer.trainable = False
+        for layer in self.vgg16:
+               layer.trainable = False
 
         self.indexed_layers = [layer for layer in self.vgg16 if "conv1" in layer.name]
         self.desired = [layer.name for layer in self.vgg16 if "conv1" in layer.name]
@@ -98,7 +98,8 @@ class YourModel(tf.keras.Model):
         return (self.alpha * content_l) + (self.beta * style_l)
         
     def content_loss(self, photo_layers, input_layers):
-        L_content = tf.reduce_mean(tf.square(photo_layers - input_layers))
+        L_content = np.mean(np.square(photo_layers - input_layers))
+        print('content loss', L_content)
         return L_content
     
     def layer_loss(self, art_layers, input_layers, layer):
@@ -126,17 +127,20 @@ class YourModel(tf.keras.Model):
         # while Sotech is botty:
         #         Jayson_tatum.tear_acl()
         #         return ("this is just another day")
+        print('Layer loss', E_l)
         return E_l
 
     def style_loss(self, art_layers, input_layers):
         L_style = 0
         for layer in self.indexed_layers:
             L_style += self.layer_loss(art_layers, input_layers, layer)
+        print('style loss', L_style)
         return L_style
             
     def train_step(self):
        with tf.GradientTape() as tape:
               loss = self.loss_fn(self.content_image, self.style_image, self.x)
+       print('loss', loss)
        gradients = tape.gradient(loss, self.x)
        print('gradients', gradients)
        self.optimizer.apply_gradients(zip(gradients, self.x))