load_model 和 Keras 中的 Lambda 层

Question

如何加载具有 lambda 层的模型？

这里是重现行为的代码：

MEAN_LANDMARKS = np.load('data/mean_shape_68.npy')

def add_mean_landmarks(x):
    mean_landmarks = np.array(MEAN_LANDMARKS, np.float32)
    mean_landmarks = mean_landmarks.flatten()
    mean_landmarks_tf = tf.convert_to_tensor(mean_landmarks)
    x = x + mean_landmarks_tf
    return x

def get_model():
    inputs = Input(shape=(8, 128, 128, 3))
    cnn = VGG16(include_top=False, weights='imagenet', input_shape=(128, 128, 3))
    x = TimeDistributed(cnn)(inputs)
    x = TimeDistributed(Flatten())(x)
    x = LSTM(256)(x)
    x = Dense(68 * 2, activation='linear')(x)

    x = Lambda(add_mean_landmarks)(x)

    model = Model(inputs=inputs, outputs=x)
    optimizer = Adadelta()
    model.compile(optimizer=optimizer, loss='mae')

    return model

模型可以编译并且我可以保存它，但是当我尝试使用 load_model 函数加载它时出现错误：

in add_mean_landmarks
    mean_landmarks = np.array(MEAN_LANDMARKS, np.float32)
NameError: name 'MEAN_LANDMARKS' is not defined

Аs 我的理解是 MEAN_LANDMARKS 没有作为常量张量包含在图中。它也与这个问题有关：

Answer 1

您需要将 custom_objects 参数传递给 load_model 函数：

model = load_model('model_file_name.h5', custom_objects={'MEAN_LANDMARKS': MEAN_LANDMARKS})

在 Keras 文档中查找更多信息：Handling custom layers (or other custom objects in saved models) .

load_model 和 Keras 中的 Lambda 层

load_model and Lamda layer in Keras

python

deep-learning

keras

tensorflow

keras-layer