Tensorflow v1.12 - ImportError: невозможно импортировать имя «облако»

Question

Я пытаюсь запустить модель тензорного потока на устройстве SBC на основе руки, близком к raspberryPi 3b +.Я скомпилировал библиотеку tenorflow из исходного кода, ориентированного на архитектуру armv7l, используя этот репозиторий https://github.com/lhelontra/tensorflow-on-arm, сохраняющий флаг поддержки GoogleCloud (GC) во время сборки.Я все еще получаю сообщение об ошибке:

  File "/home/pi/tensorflow_pi_test/predict_image_v6.py", line 165, in <module>
    result = predict_image_tflite('./1.jpg', './denseNet201_model_final.pb')
  File "/home/pi/tensorflow_pi_test/predict_image_v6.py", line 143, in predict_image_tflite
    interpreter = tf.contrib.lite.Interpreter(model_path="converted_model.tflite")
  File "/home/pi/venv_tf_1.10/lib/python3.5/site-packages/tensorflow/python/util/lazy_loader.py", line 53, in __getattr__
    module = self._load()
  File "/home/pi/venv_tf_1.10/lib/python3.5/site-packages/tensorflow/python/util/lazy_loader.py", line 42, in _load
    module = importlib.import_module(self.__name__)
  File "/home/pi/venv_tf_1.10/lib/python3.5/importlib/__init__.py", line 126, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
  File "<frozen importlib._bootstrap>", line 986, in _gcd_import
  File "<frozen importlib._bootstrap>", line 969, in _find_and_load
  File "<frozen importlib._bootstrap>", line 958, in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 673, in _load_unlocked
  File "<frozen importlib._bootstrap_external>", line 673, in exec_module
  File "<frozen importlib._bootstrap>", line 222, in _call_with_frames_removed
  File "/home/pi/venv_tf_1.10/lib/python3.5/site-packages/tensorflow/contrib/__init__.py", line 38, in <module>
    from tensorflow.contrib import cloud
ImportError: cannot import name 'cloud'

Process finished with exit code 1

Код, который я использую для выполнения модели на платформе arm:

def predict_image_tflite(image_path, model_path):
    short_edge_min = 256
    short_edge_max = 384
    center_crop_size = (224, 224)

    # open image
    img = Image.open(image_path)
    #     show image
    #     plt.imshow(img)

    # downsample image
    width, height = img.size
    tmp = min(width, height)
    short_edge_resize = np.random.randint(short_edge_min, short_edge_max)

    # check python version
    py_ver = platform.python_version()

    if py_ver.split('.')[0] == '3':
        # python 3
        width = int(width * short_edge_resize / tmp)
        height = int(height * short_edge_resize / tmp)
    elif py_ver.split('.')[0] == '2':
        # python 2
        width = width * short_edge_resize / tmp
        height = height * short_edge_resize / tmp

    img = img.resize((width, height))
    img = np.array(img)

    # center crop image
    centerw, centerh = img.shape[1] // 2, img.shape[0] // 2
    halfw, halfh = center_crop_size[1] // 2, center_crop_size[0] // 2

    offsetw, offseth = 0, 0
    if center_crop_size[0] % 2 == 1:
        offseth = 1
    if center_crop_size[1] % 2 == 1:
        offsetw = 1

    img = img[centerh - halfh:centerh + halfh + offseth, centerw - halfw:centerw + halfw + offsetw, :]

    # expand image dimension to 4D
    img = np.expand_dims(img, axis=0)


    ## Now the prediction script runs
    # Load TFLite model and allocate tensors.
    interpreter = tf.contrib.lite.Interpreter(model_path="converted_model.tflite")
    interpreter.allocate_tensors()

    # Get input and output tensors.
    input_details = interpreter.get_input_details()
    output_details = interpreter.get_output_details()

    #Testing out model on random test data

    input_shape = input_details[0]['shape']
    input_data = np.array(np.random.random_sample(input_shape), dtype=np.float32)
    interpreter.set_tensor(input_details[0]['index'], input_data)


    interpreter.invoke()
    output_data = interpreter.get_tensor(output_details[0]['index'])
    print(input_data)
    print(output_data)

Есть предложения о том, как подойти к этому блокировщику?