Keras遷移學習實現影象分類和特徵提取
阿新 • • 發佈:2018-12-21
Kera的應用模組Application提供了帶有預訓練權重的Keras模型,這些模型可以用來進行預測、特徵提取和finetune
模型的預訓練權重將下載到~/.keras/models/
並在載入模型時自動載入
可用的模型
所有的這些模型(除了Xception和MobileNet)都相容Theano和Tensorflow,並會自動基於~/.keras/keras.json
的Keras的影象維度進行自動設定。例如,如果你設定data_format="channel_last"
,則載入的模型將按照TensorFlow的維度順序來構造,即“Width-Height-Depth”的順序
Xception模型僅在TensorFlow下可用,因為它依賴的SeparableConvolution層僅在TensorFlow可用。MobileNet僅在TensorFlow下可用,因為它依賴的DepethwiseConvolution層僅在TF下可用。
以上模型(暫時除了MobileNet)的預訓練權重可以在我的百度網盤下載,如果有更新的話會在這裡報告
圖片分類模型的示例
利用ResNet50網路進行ImageNet分類
from keras.applications.resnet50 import ResNet50 from keras.preprocessing import image from keras.applications.resnet50 import preprocess_input, decode_predictions import numpy as np model = ResNet50(weights='imagenet') img_path = 'elephant.jpg' img = image.load_img(img_path, target_size=(224, 224)) x = image.img_to_array(img) x = np.expand_dims(x, axis=0) x = preprocess_input(x) preds = model.predict(x) # decode the results into a list of tuples (class, description, probability) # (one such list for each sample in the batch) print('Predicted:', decode_predictions(preds, top=3)[0]) # Predicted: [(u'n02504013', u'Indian_elephant', 0.82658225), (u'n01871265', u'tusker', 0.1122357), (u'n02504458', u'African_elephant', 0.061040461)]
利用VGG16提取特徵
from keras.applications.vgg16 import VGG16 from keras.preprocessing import image from keras.applications.vgg16 import preprocess_input import numpy as np model = VGG16(weights='imagenet', include_top=False) img_path = 'elephant.jpg' img = image.load_img(img_path, target_size=(224, 224)) x = image.img_to_array(img) x = np.expand_dims(x, axis=0) x = preprocess_input(x) features = model.predict(x)
從VGG19的任意中間層中抽取特徵
from keras.applications.vgg19 import VGG19
from keras.preprocessing import image
from keras.applications.vgg19 import preprocess_input
from keras.models import Model
import numpy as np
base_model = VGG19(weights='imagenet')
model = Model(inputs=base_model.input, outputs=base_model.get_layer('block4_pool').output)
img_path = 'elephant.jpg'
img = image.load_img(img_path, target_size=(224, 224))
x = image.img_to_array(img)
x = np.expand_dims(x, axis=0)
x = preprocess_input(x)
block4_pool_features = model.predict(x)
在新類別上fine-tune inceptionV3
from keras.applications.inception_v3 import InceptionV3
from keras.preprocessing import image
from keras.models import Model
from keras.layers import Dense, GlobalAveragePooling2D
from keras import backend as K
# 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(200, activation='softmax')(x)
# this is the model we will train
model = Model(inputs=base_model.input, outputs=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(...)
# 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 249 layers and unfreeze the rest:
for layer in model.layers[:249]:
layer.trainable = False
for layer in model.layers[249:]:
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(...)
在定製的輸入tensor上構建InceptionV3
from keras.applications.inception_v3 import InceptionV3
from keras.layers import Input
# this could also be the output a different Keras model or layer
input_tensor = Input(shape=(224, 224, 3)) # this assumes K.image_data_format() == 'channels_last'
model = InceptionV3(input_tensor=input_tensor, weights='imagenet', include_top=True)
模型資訊
模型 | 大小 | Top1準確率 | Top5準確率 | 引數數目 | 深度 |
---|---|---|---|---|---|
Xception | 88MB | 0.790 | 0.945 | 22,910,480 | 126 |
VGG16 | 528MB | 0.715 | 0.901 | 138,357,544 | 23 |
VGG19 | 549MB | 0.727 | 0.910 | 143,667,240 | 26 |
ResNet50 | 99MB | 0.759 | 0.929 | 25,636,712 | 168 |
InceptionV3 | 92MB | 0.788 | 0.944 | 23,851,784 | 159 |
IncetionResNetV2 | 215MB | 0.804 | 0.953 | 55,873,736 | 572 |
MobileNet | 17MB | 0.665 | 0.871 | 4,253,864 | 88 |
MobileNetV2 | 14MB | 0.713 | 0.901 | 3,538,984 | 88 |
DenseNet121 | 33MB | 0.750 | 0.923 | 8,062,504 | 121 |
DenseNet169 | 57MB | 0.762 | 0.932 | 14,307,880 | 169 |
DenseNet201 | 80MB | 0.773 | 0.936 | 20,242,984 | 201 |
NASNetMobile | 23MB | 0.744 | 0.919 | 5,326,716 | - |
NASNetLarge | 343MB | 0.825 | 0.960 | 88,949,818 | - |