我正在尝试使用在Keras中预先训练的VGG16,但我一直收到此错误:
ValueError:检查目标时出错:预期block5_pool具有4维,但数组的形状为(6,1)
错误是什么意思?
检查目标时出错:期望block5_pool具有4个维度, 但是得到形状为(6,1)的数组
这是我的实际代码:
from keras import applications
from keras.preprocessing.image import ImageDataGenerator
from keras.models import Sequential
from keras.layers import Conv2D, MaxPooling2D, ZeroPadding2D
from keras.layers import Activation, Dropout, Flatten, Dense
from keras import backend as K
# dimensions of our images.
img_width, img_height = 224, 224
train_data_dir = 'database/train'
validation_data_dir = 'database/validation'
nb_train_samples = 2000
nb_validation_samples = 26
epochs = 50
batch_size = 20
if K.image_data_format() == 'channels_first':
input_shape = (3, img_width, img_height)
else:
input_shape = (img_width, img_height, 3)
model = applications.VGG16(weights='imagenet', include_top=False, input_shape=(224, 224, 3))
print('VGG-16 Model loaded.')
top_model = Sequential()
top_model.add(ZeroPadding2D((1,1),input_shape=input_shape))
top_model.add(Conv2D(64, (3, 3), activation='relu'))
top_model.add(ZeroPadding2D((1,1)))
top_model.add(Conv2D(64, (3, 3), activation='relu'))
top_model.add(MaxPooling2D(pool_size=(2,2), strides=(2,2)))
top_model.add(ZeroPadding2D((1,1)))
top_model.add(Conv2D(128, (3, 3), activation='relu'))
top_model.add(ZeroPadding2D((1,1)))
top_model.add(Conv2D(128, (3, 3), activation='relu'))
top_model.add(MaxPooling2D((2,2), strides=(2,2)))
top_model.add(ZeroPadding2D((1,1)))
top_model.add(Conv2D(256, (3, 3), activation='relu'))
top_model.add(ZeroPadding2D((1,1)))
top_model.add(Conv2D(256, (3, 3), activation='relu'))
top_model.add(ZeroPadding2D((1,1)))
top_model.add(Conv2D(256, (3, 3), activation='relu'))
top_model.add(MaxPooling2D((2,2), strides=(2,2)))
top_model.add(ZeroPadding2D((1,1)))
top_model.add(Conv2D(512, (3, 3), activation='relu'))
top_model.add(ZeroPadding2D((1,1)))
top_model.add(Conv2D(512, (3, 3), activation='relu'))
top_model.add(ZeroPadding2D((1,1)))
top_model.add(Conv2D(512, (3, 3), activation='relu'))
top_model.add(MaxPooling2D((2,2), strides=(2,2)))
top_model.add(ZeroPadding2D((1,1)))
top_model.add(Conv2D(512, (3, 3), activation='relu'))
top_model.add(ZeroPadding2D((1,1)))
top_model.add(Conv2D(512, (3, 3), activation='relu'))
top_model.add(ZeroPadding2D((1,1)))
top_model.add(Conv2D(512, (3, 3), activation='relu'))
top_model.add(MaxPooling2D((2,2), strides=(2,2)))
top_model.add(Flatten())
top_model.add(Dense(4096, activation='relu'))
top_model.add(Dropout(0.5))
top_model.add(Dense(12))
top_model.add(Activation('softmax'))
# note that it is necessary to start with a fully-trained
# classifier, including the top classifier,
# in order to successfully do fine-tuning
# top_model.load_weights('./vgg16_face_weights.h5')
# add the model on top of the convolutional base
model.add_update(top_model)
# set the first 25 layers (up to the last conv block)
# to non-trainable (weights will not be updated)
for layer in model.layers[:25]:
layer.trainable = False
model.compile(loss='sparse_categorical_crossentropy', optimizer='rmsprop', metrics=['accuracy'])
# this is the augmentation configuration we will use for training
train_datagen = ImageDataGenerator(
rescale=1. / 255,
shear_range=0.2,
zoom_range=0.2,
horizontal_flip=True)
# this is the augmentation configuration we will use for testing:
# only rescaling
test_datagen = ImageDataGenerator(rescale=1. / 255)
train_generator = train_datagen.flow_from_directory(
train_data_dir,
target_size=(img_width, img_height),
batch_size=batch_size,
class_mode='binary')
validation_generator = test_datagen.flow_from_directory(
validation_data_dir,
target_size=(img_width, img_height),
batch_size=batch_size,
class_mode='binary')
model.fit_generator(
train_generator,
steps_per_epoch=nb_train_samples // batch_size,
epochs=epochs,
validation_data=validation_generator,
validation_steps=nb_validation_samples // batch_size)
model.save_weights('first_try.h5')
终端运行:
VGG-16 Model loaded.
Found 46 images belonging to 12 classes.
Found 26 images belonging to 12 classes.
Epoch 1/50
Traceback (most recent call last):
File "C:/Users/w024029h/PycharmProjects/keras_pretrained/pretrained.py", line 113, in <module>
validation_steps=nb_validation_samples // batch_size)
File "C:\Users\w024029h\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\legacy\interfaces.py", line 91, in wrapper
return func(*args, **kwargs)
File "C:\Users\w024029h\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\engine\training.py", line 2230, in fit_generator
class_weight=class_weight)
File "C:\Users\w024029h\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\engine\training.py", line 1877, in train_on_batch
class_weight=class_weight)
File "C:\Users\w024029h\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\engine\training.py", line 1480, in _standardize_user_data
exception_prefix='target')
File "C:\Users\w024029h\AppData\Local\Programs\Python\Python36\lib\site-packages\keras\engine\training.py", line 113, in _standardize_input_data
'with shape ' + str(data_shape))
ValueError: Error when checking target: expected block5_pool to have 4 dimensions, but got array with shape (6, 1)
答案 0 :(得分:2)
我之前也遇到过同样的错误。我通过将数据生成器上的class_mode从'binary'更改为None
来解决了该问题根据keras文档,将class_mode设置为“ binary”将返回一维数组。
通过更改class_mode = None,不会返回任何标签,并且生成器将仅生成一批图像数据。这就是您对模型的期望,并且可以与fit_generator()方法配合使用。但是,当使用class_mode None时,数据仍需要驻留在目录的子目录中才能正常工作。
^看看flow_from_directory()
希望这会有所帮助!