无法通过第一个时代--只是挂起[Keras传输学习初始阶段]

Question

我基本上是在使用传输学习API教程中的大部分代码，

https://faroit.github.io/keras-docs/2.0.0/applications/#inceptionv3

只是做了几处小小的改动来适应我的数据。

我正在使用Tensorflow-GPU1.4、Windows 7和Keras 2.03(？最新的Keras)。

代码：

from keras.applications.inception_v3 import InceptionV3
from keras.preprocessing import image
from keras.preprocessing.image import ImageDataGenerator
from keras.models import Model
from keras.layers import Dense, GlobalAveragePooling2D
from keras import backend as K


img_width, img_height = 299, 299
train_data_dir = r'C:\Users\Moondra\Desktop\Keras Applications\data\train'
nb_train_samples = 8
nb_validation_samples = 100 
batch_size = 10
epochs = 5


train_datagen = ImageDataGenerator(
rescale = 1./255,
horizontal_flip = True,
zoom_range = 0.1,
rotation_range=15)



train_generator = train_datagen.flow_from_directory(
train_data_dir,
target_size = (img_height, img_width),
batch_size = batch_size, 
class_mode = 'categorical')  #class_mode = 'categorical'


# create the base pre-trained model
base_model = InceptionV3(weights='imagenet', include_top=False)

# add a global spatial average pooling layer
x = base_model.output
x = GlobalAveragePooling2D()(x)
# let's add a fully-connected layer
x = Dense(1024, activation='relu')(x)
# and a logistic layer -- let's say we have 200 classes
predictions = Dense(12, activation='softmax')(x)

# this is the model we will train
model = Model(input=base_model.input, output=predictions)

# first: train only the top layers (which were randomly initialized)
# i.e. freeze all convolutional InceptionV3 layers
for layer in base_model.layers:
    layer.trainable = False

# compile the model (should be done *after* setting layers to non-trainable)
model.compile(optimizer='rmsprop', loss='categorical_crossentropy')

# train the model on the new data for a few epochs
model.fit_generator(
train_generator,
steps_per_epoch = 5,
epochs = epochs)


# at this point, the top layers are well trained and we can start fine-tuning
# convolutional layers from inception V3. We will freeze the bottom N layers
# and train the remaining top layers.

# let's visualize layer names and layer indices to see how many layers
# we should freeze:
for i, layer in enumerate(base_model.layers):
   print(i, layer.name)

# we chose to train the top 2 inception blocks, i.e. we will freeze
# the first 172 layers and unfreeze the rest:
for layer in model.layers[:172]:
   layer.trainable = False
for layer in model.layers[172:]:
   layer.trainable = True

# we need to recompile the model for these modifications to take effect
# we use SGD with a low learning rate
from keras.optimizers import SGD
model.compile(optimizer=SGD(lr=0.0001, momentum=0.9), loss='categorical_crossentropy')

# we train our model again (this time fine-tuning the top 2 inception blocks
# alongside the top Dense layers
model.fit_generator(
train_generator,
steps_per_epoch = 5,
epochs = epochs)

输出(过不了第一个时代)：

 Epoch 1/5

1/5 [=====>........................] - ETA: 8s - loss: 2.4869
2/5 [===========>..................] - ETA: 3s - loss: 5.5591
3/5 [=================>............] - ETA: 1s - loss: 6.6299

4/5 [=======================>......] - ETA: 0s - loss: 8.4925

它就挂在这里。

更新：

我用tensorflow 1.3 (降级一个版本)和Keras 2.03(最新的pip版本)创建了一个虚拟env，但仍然存在相同的问题。

更新2

我不认为这是一个记忆问题，就好像我改变了时代中的步骤一样--它会一直运行到最后一步，然后冻结。

所以在一个时代里走30步，它就会持续到29步。

5步，它会一直跑到第4步，然后挂起来。

更新3

还尝试了Keras中建议的249层。

Answer 1

很明显，这是一个通过Keras更新而得到修复的错误(然而，一些人仍然在经历这个问题)。

Answer 2

似乎大多数冻结问题都发生在代码中的某些错误发生时。在我的案子中，我构建了一个生成器，它在划时代结束时抛出一个异常，然后进程停止。但是没有关于异常的信息，所以我也花了一些时间来弄清楚到底是怎么回事。

Answer 3

正如@thomas所提到的，我在keras/tf兼容性方面也有类似的问题。具体来说，我的配置是：cuda-10.0, cudnn-7, tensorflow_gpu=1.14.0, keras=2.2.5.

修正了将其降级为：cuda-9.0, cudnn-7, tensorflow-gpu=1.10.0 and keras=2.2.0

从本文中得到了关于不兼容性的想法：https://github.com/tensorflow/tensorflow/issues/15604

此外，您还可以在以下文章中引用keras和tensorflow兼容性：

1. 配置
2. https://docs.floydhub.com/guides/environments/