Update app.py

app.py

@@ -12,10 +12,7 @@ autoencoder = load_model("autoencoder_model.keras")
 encoded_images = np.load("X_encoded_compressed.npy")
 print("Shape of encoded_images:", encoded_images.shape)
 print("Sample encoded image:", encoded_images[0])
-dataset = load_dataset('eybro/images')
-split_dataset = dataset['train'].train_test_split(test_size=0.2)
-dataset['train'] = split_dataset['train']
-dataset['test'] = split_dataset['test']
+
 
 num_clusters = 10  # Choose the number of clusters
 kmeans = KMeans(n_clusters=num_clusters, random_state=42)
@@ -41,11 +38,15 @@ def find_nearest_neighbors(encoded_images, input_image, top_n=5):
     return [(index, distances[index]) for index in nearest_neighbors]
 
 def get_image(index):
-  split = len(dataset["train"])
-  if index < split:
-    return dataset["train"][index]
-  else:
-    return dataset["test"][index-split]
+    dataset = load_dataset('eybro/images')
+    split_dataset = dataset['train'].train_test_split(test_size=0.2)
+    dataset['train'] = split_dataset['train']
+    dataset['test'] = split_dataset['test']
+    split = len(dataset["train"])
+    if index < split:
+        return dataset["train"][index]
+    else:
+        return dataset["test"][index-split]
 
 def process_image(image):
     img = np.array(image)
@@ -82,8 +83,8 @@ def inference(image):
       im = get_image(i)
       print(im["label"], im["timestamp"])
 
-    result_image = get_image(top4[0])
-    result = f"{result_image['label']} {result_image['timestamp']}"
+    #result_image = get_image(top4[0])
+    #result = f"{result_image['label']} {result_image['timestamp']}"
 
     n=2
     plt.figure(figsize=(8, 8))
@@ -103,7 +104,7 @@ def inference(image):
         ax.get_xaxis().set_visible(False)
         ax.get_yaxis().set_visible(False)
     
-    return result
+    return None