Update app.py

app.py

@@ -25,14 +25,23 @@ model = tf.keras.models.load_model(
 
 # 3.prediction function (predict())
 
-def load_and_prep_imgg(filename, img_shape=224, scale=True):
+def load_and_prep_imgg(img : Image.Image, img_shape=224, scale=True):
     if not isinstance(filename, str):
         raise ValueError("The filename must be a string representing the file path.")
-    img = tf.io.read_file(filename)
-    img = tf.io.decode_image(img, channels=3)
-    img = tf.image.resize(img, size=[img_shape, img_shape])
+    # img = tf.io.read_file(filename)
+    # img = tf.io.decode_image(img, channels=3)
+    # img = tf.image.resize(img, size=[img_shape, img_shape])
+    # if scale:
+    #     return img / 255
+    # else:
+    #     return img
+    img = img.resize((img_shape, img_shape))
+    img = np.array(img)
+    if img.shape[-1] == 1:  # If the image is grayscale
+        img = np.stack([img] * 3, axis=-1)
+    img = tf.convert_to_tensor(img, dtype=tf.float32)
     if scale:
-        return img / 255
+        return img / 255.0
     else:
         return img
         
@@ -57,10 +66,9 @@ def predict(img) -> Tuple[Dict,float] :
 
 ### 4. Gradio app - our Gradio interface + launch command
 
-title = 'FoodVision Big'
-description = 'Feature Extraxtion VGG  model to classifiy Macular Diseases by OCT '
-article = 'created at Tensorflow Model Deployment'
-
+title = 'Macular Disease Classification'
+description = 'Feature Extraction VGG model to classify Macular Diseases by OCT'
+article = 'Created with TensorFlow Model Deployment'
 # Create example list
 
 example_list = [['examples/'+ example] for example in os.listdir('examples')]