remove model monitoring

app.py

@@ -31,76 +31,7 @@ N_IMAGES = 100
 # Define the model and its parameters
 #model = joblib.load(filename="plant_leaf_diseases_model.keras", 'utf-8')
 model = tf.keras.models.load_model('plant_leaf_diseases_model.keras')
-#test_dir_img_file_path = os.path.getcwd("imagesfolder")
-test_dir_img_file_path = os.path.join(os.getcwd(),"imagesfolder")
-# Define the metrics
-accuracy = prom.Gauge("accuracy", "precision score for random 50 test samples")
-precision_metric = prom.Gauge("plant_precision", "precision score for random 50 test samples")
-recall_metric = prom.Gauge("plant_recall_score", "recall score for random 50 test samples")
-f1_metric = prom.Gauge("plant_f1_score", "F1 score for random 50 test samples")
-
-# Define the function to make predictions
-#def make_prediction(test_dir_img_file_path):
-#    # Load the test images
-#    image_list = []
-#    for plant_image in os.listdir(test_dir_img_file_path):
-##        if plant_image.endswith(".jpg") or plant_image.endswith(".JPG"):
-#            image_list.append(convert_image_to_array(os.path.join(test_dir_img_file_path, plant_image)))
-#    np_image_list = np.array(image_list, dtype=np.float16) / config.model_config.scaling_factor
-#
-#    # Make predictions
-#    predictions = model.predict(np_image_list, verbose=0)
-
-    # Return the predictions
-#    return {"predictions": predictions, "version": _version}
-def convert_image_to_array(image_dir):
-    try:
-        print(image_dir)
-        image = cv2.imread(image_dir)
-        if image is not None:
-            image = cv2.resize(image, DEFAULT_IMAGE_SIZE)
-            return img_to_array(image)
-        else :
-            return np.array([])
-    except Exception as e:
-        print(f"Error : {e}")
-        return None
-# Define the function to update the metrics
-x_test, y_test = train_test_split(test_dir_img_file_path)
-def update_metrics():
-    # Load the test images
-    image_list = []
-    plant_disease_folder_list = listdir(test_dir_img_file_path)
-    for plant_disease_folder in plant_disease_folder_list:
-        plant_disease_image_list = listdir(f"{test_dir_img_file_path}/{plant_disease_folder}/")
-        for plant_image in plant_disease_image_list[:N_IMAGES]:
-            plant_image_directory = f"{test_dir_img_file_path}/{plant_disease_folder}/{image}"
-        if plant_image_directory.endswith(".jpg") or plant_image_directory.endswith(".JPG"):
-            image_list.append(convert_image_to_array(os.path.join(plant_image_directory)))
-            label_list.append(plant_disease_folder)
-            image_len = len(image_list)
-            print(f"Total number of images: {image_len}")
-    np_image_list = np.array(image_list, dtype=np.float16) / 255.0
-    # Make predictions
-    predictions = model.predict(np_image_list, verbose=0)
-
-    # Calculate the metrics
-    pred = np.argmax(predictions, axis=1)
-    y_test2 = np.argmax(y_test, axis=1)
-    accuracy = accuracy_score(y_true, y_pred)
-    precision = precision_score(y_test2, pred)
-    recall = recall_score(y_test2, predictions)
-    f1 = f1_score(y_test2, predictions)
-
-    # Update the metrics
-    precision_metric.set(precision)
-    recall_metric.set(recall)
-    f1_metric.set(f1)
-
-@app.get("/metrics")
-async def get_metrics():
-    update_metrics()
-    return Response(media_type="text/plain", content= prom.generate_latest())
+
 
 
 myControls = {