Question

我正在开发一个简单的keras模型，我想使用一些回调，例如Tensorboard和ReduceLROnPlateau，但是出现类似以下错误：

tensorflow.python.framework.errors_impl.FailedPreconditionError：尝试使用未初始化的值training / Adam / Variable_81

模型可在tensorflow 1.14，python 3.6，keras 2.2.4中使用

我尝试了keras.backends.get_session（）。run（*** initializers），但是仍然显示相同的错误。

log_dir  ='logs/'
data_gen_args = dict(rotation_range=0.2,
                width_shift_range=0.05,
                height_shift_range=0.05,
                shear_range=0.05,
                zoom_range=0.05,
                horizontal_flip=True,
                fill_mode='nearest')  

myGene=trainGenerator(2,'data/membrane/train','image','label',data_gen_args,save_to_dir = None)
model_checkpoint = ModelCheckpoint(log_dir + "ep{epoch:03d}-loss{loss:.3f}.h5",
    monitor='loss', save_weights_only=True, save_best_only=True, period=1)
logging = TensorBoard(log_dir=log_dir)
reduce_lr = ReduceLROnPlateau(monitor='val_loss', factor=0.1, patience=3, verbose=1)

early_stopping = EarlyStopping(monitor='loss', min_delta=0, patience=3, verbose=1)

model = unet()

model.fit_generator(myGene,steps_per_epoch=5,epochs=10,callbacks=[logging, reduce_lr, early_stopping, model_checkpoint])

##########################
model codes, another .py file
##########################

from keras.models import *

from keras.layers import *

from keras.optimizers import *

from keras.callbacks import ModelCheckpoint, ReduceLROnPlateau

from keras import backend as keras


def unet(pretrained_weights = None,input_size = (256,256,1)):
    inputs = Input(input_size)
    conv1
    ......
    conv10 = Conv2D(1, 1, activation = 'sigmoid')(conv9)

    model = Model(input = inputs, output = conv10)

    model.compile(optimizer = Adam(lr = 1e-4), loss = 
    'binary_crossentropy', metrics = ['accuracy'])

    return model

early_stopping回调工作良好，也许与Adam无关。我只想在keras模型中添加一些回调函数。

或者与model.fit_generator有关系吗？

FailedPreconditionError：尝试使用未初始化的值训练/ Adam / *

0 个答案: