Update model.py

model.py

@@ -280,25 +280,25 @@ def get_caption_model():
 
     cnn_model = CNN_Encoder()
 
-    caption_mode = ImageCaptioningModel(
+    caption_model = ImageCaptioningModel(
         cnn_model=cnn_model, encoder=encoder, decoder=decoder, image_aug=None,
     )
 
     def call_fn(batch, training):
         return batch
 
-    caption_mode.call = call_fn
+    caption_model.call = call_fn
     sample_x, sample_y = tf.random.normal((1, 299, 299, 3)), tf.zeros((1, 40))
 
-    caption_mode((sample_x, sample_y))
+    caption_model((sample_x, sample_y))
 
-    sample_img_embed = caption_mode.cnn_model(sample_x)
-    sample_enc_out = caption_mode.encoder(sample_img_embed, training=False)
-    caption_mode.decoder(sample_y, sample_enc_out, training=False)
+    sample_img_embed = caption_model.cnn_model(sample_x)
+    sample_enc_out = caption_model.encoder(sample_img_embed, training=False)
+    caption_model.decoder(sample_y, sample_enc_out, training=False)
 
-    caption_mode.load_weights('model.h5')
+    caption_model.load_weights('model.h5')
 
-    return caption_mode
+    return caption_model
 
 def load_image_from_path(img_path):
     img = tf.io.read_file(img_path)
@@ -317,14 +317,14 @@ def generate_caption(img_path, add_noise=False):
         img = (img - tf.reduce_min(img))/(tf.reduce_max(img) - tf.reduce_min(img))
     
     img = tf.expand_dims(img, axis=0)
-    img_embed = model.cnn_model(img)
-    img_encoded = model.encoder(img_embed, training=False)
+    img_embed = caption_model.cnn_model(img)
+    img_encoded = caption_model.encoder(img_embed, training=False)
 
     y_inp = '[start]'
     for i in range(MAX_LENGTH-1):
         tokenized = tokenizer([y_inp])[:, :-1]
         mask = tf.cast(tokenized != 0, tf.int32)
-        pred = model.decoder(
+        pred = caption_model.decoder(
             tokenized, img_encoded, training=False, mask=mask)
         
         pred_idx = np.argmax(pred[0, i, :])