在 Keras 中训练 multi-class image classifier
Train multi-class image classifier in Keras
我正在学习使用 Keras
训练 classifier 的教程
https://blog.keras.io/building-powerful-image-classification-models-using-very-little-data.html
具体来说，根据作者给出的 second script，我想将脚本转换成一个可以训练 multi-class classifier 的脚本（是 cat 的二进制文件和狗）。我的 train 文件夹中有 5 classes，所以我做了以下更改：
在train_top_model()的函数中：
我变了
model = Sequential()
model.add(Flatten(input_shape=train_data.shape[1:]))
model.add(Dense(256, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(1, activation='sigmoid'))
model.compile(optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])
进入
model = Sequential()
model.add(Flatten(input_shape=train_data.shape[1:]))
model.add(Dense(256, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(5, activation='sigmoid'))
model.compile(optimizer='rmsprop', loss='categorical_crossentropy', metrics=['accuracy'])
train_labels = to_categorical(train_labels, 5)
validation_labels = to_categorical(validation_labels, 5)
完成训练后，模型达到了接近 99% 的训练准确率，但只有验证准确率的 70% 左右。因此我开始思考，也许将 2 classes 训练转换为 5 classes 并不是那么简单。也许我需要在标记 classes 时使用 one-hot 编码（但我不知道如何）
编辑：
我也附上了我的微调脚本。另一个问题：fine-tuning开始时精度没有有效提升
import os
import h5py
import numpy as np
from keras.preprocessing.image import ImageDataGenerator
from keras import optimizers
from keras.models import Sequential
from keras.layers import Convolution2D, MaxPooling2D, ZeroPadding2D
from keras.layers import Activation, Dropout, Flatten, Dense
# path to the model weights files.
weights_path = 'D:/Users/EJLTZ/Desktop/vgg16_weights.h5'
top_model_weights_path = 'bottleneck_weights_2.h5'
# dimensions of our images.
img_width, img_height = 150, 150
train_data_dir = 'D:/Users/EJLTZ/Desktop/BodyPart-full/train_new'
validation_data_dir = 'D:/Users/EJLTZ/Desktop/BodyPart-full/validation_new'
nb_train_samples = 500
nb_validation_samples = 972
nb_epoch = 50
# build the VGG16 network
model = Sequential()
model.add(ZeroPadding2D((1, 1), input_shape=(3, img_width, img_height)))
model.add(Convolution2D(64, 3, 3, activation='relu', name='conv1_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(64, 3, 3, activation='relu', name='conv1_2'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(128, 3, 3, activation='relu', name='conv2_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(128, 3, 3, activation='relu', name='conv2_2'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(256, 3, 3, activation='relu', name='conv3_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(256, 3, 3, activation='relu', name='conv3_2'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(256, 3, 3, activation='relu', name='conv3_3'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv4_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv4_2'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv4_3'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv5_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv5_2'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv5_3'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
# load the weights of the VGG16 networks
# (trained on ImageNet, won the ILSVRC competition in 2014)
# note: when there is a complete match between your model definition
# and your weight savefile, you can simply call model.load_weights(filename)
assert os.path.exists(weights_path), 'Model weights not found (see "weights_path" variable in script).'
f = h5py.File(weights_path)
for k in range(f.attrs['nb_layers']):
if k >= len(model.layers):
# we don't look at the last (fully-connected) layers in the savefile
break
g = f['layer_{}'.format(k)]
weights = [g['param_{}'.format(p)] for p in range(g.attrs['nb_params'])]
model.layers[k].set_weights(weights)
f.close()
print('Model loaded.')
# build a classifier model to put on top of the convolutional model
top_model = Sequential()
top_model.add(Flatten(input_shape=model.output_shape[1:]))
top_model.add(Dense(256, activation='relu'))
top_model.add(Dropout(0.5))
top_model.add(Dense(5, 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(top_model_weights_path)
# add the model on top of the convolutional base
model.add(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
# compile the model with a SGD/momentum optimizer
# and a very slow learning rate.
model.compile(loss='categorical_crossentropy',
optimizer=optimizers.SGD(lr=1e-4, momentum=0.9),
metrics=['accuracy'])
# prepare data augmentation configuration
train_datagen = ImageDataGenerator(
rescale=1./255,
shear_range=0.2,
zoom_range=0.2,
horizontal_flip=True)
test_datagen = ImageDataGenerator(rescale=1./255)
train_generator = train_datagen.flow_from_directory(
train_data_dir,
target_size=(img_height, img_width),
batch_size=32,
class_mode= 'categorical')
validation_generator = test_datagen.flow_from_directory(
validation_data_dir,
target_size=(img_height, img_width),
batch_size=32,
class_mode= 'categorical')
# fine-tune the model
model.fit_generator(
train_generator,
samples_per_epoch=nb_train_samples,
nb_epoch=nb_epoch,
validation_data=validation_generator,
nb_val_samples=nb_validation_samples)
model.save_weights("fine-tune_weights.h5")
model.save("fine-tune_model.h5", True)
使用softmax作为输出层的激活函数，它是logistic函数在多class情况下的推广。阅读更多相关信息 here.
如果验证误差比训练误差大得多，如您的情况，则表明存在过度拟合。你应该做一些正则化，它被定义为学习算法的任何变化，旨在减少测试错误而不是训练错误。您可以尝试诸如数据增强、提前停止、噪声注入、更积极的 dropout 等方法
如果您有与链接教程中相同的 set-up，请将 train_generator 和 validation_generator 的 class_mode 更改为 categorical 并且它将 one-hot 编码你的 classes。
我正在学习使用 Keras
训练 classifier 的教程https://blog.keras.io/building-powerful-image-classification-models-using-very-little-data.html
具体来说，根据作者给出的 second script，我想将脚本转换成一个可以训练 multi-class classifier 的脚本（是 cat 的二进制文件和狗）。我的 train 文件夹中有 5 classes，所以我做了以下更改：
在train_top_model()的函数中：
我变了
model = Sequential()
model.add(Flatten(input_shape=train_data.shape[1:]))
model.add(Dense(256, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(1, activation='sigmoid'))
model.compile(optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])
进入
model = Sequential()
model.add(Flatten(input_shape=train_data.shape[1:]))
model.add(Dense(256, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(5, activation='sigmoid'))
model.compile(optimizer='rmsprop', loss='categorical_crossentropy', metrics=['accuracy'])
train_labels = to_categorical(train_labels, 5)
validation_labels = to_categorical(validation_labels, 5)
完成训练后，模型达到了接近 99% 的训练准确率，但只有验证准确率的 70% 左右。因此我开始思考，也许将 2 classes 训练转换为 5 classes 并不是那么简单。也许我需要在标记 classes 时使用 one-hot 编码（但我不知道如何）
编辑：
我也附上了我的微调脚本。另一个问题：fine-tuning开始时精度没有有效提升
import os
import h5py
import numpy as np
from keras.preprocessing.image import ImageDataGenerator
from keras import optimizers
from keras.models import Sequential
from keras.layers import Convolution2D, MaxPooling2D, ZeroPadding2D
from keras.layers import Activation, Dropout, Flatten, Dense
# path to the model weights files.
weights_path = 'D:/Users/EJLTZ/Desktop/vgg16_weights.h5'
top_model_weights_path = 'bottleneck_weights_2.h5'
# dimensions of our images.
img_width, img_height = 150, 150
train_data_dir = 'D:/Users/EJLTZ/Desktop/BodyPart-full/train_new'
validation_data_dir = 'D:/Users/EJLTZ/Desktop/BodyPart-full/validation_new'
nb_train_samples = 500
nb_validation_samples = 972
nb_epoch = 50
# build the VGG16 network
model = Sequential()
model.add(ZeroPadding2D((1, 1), input_shape=(3, img_width, img_height)))
model.add(Convolution2D(64, 3, 3, activation='relu', name='conv1_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(64, 3, 3, activation='relu', name='conv1_2'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(128, 3, 3, activation='relu', name='conv2_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(128, 3, 3, activation='relu', name='conv2_2'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(256, 3, 3, activation='relu', name='conv3_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(256, 3, 3, activation='relu', name='conv3_2'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(256, 3, 3, activation='relu', name='conv3_3'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv4_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv4_2'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv4_3'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv5_1'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv5_2'))
model.add(ZeroPadding2D((1, 1)))
model.add(Convolution2D(512, 3, 3, activation='relu', name='conv5_3'))
model.add(MaxPooling2D((2, 2), strides=(2, 2)))
# load the weights of the VGG16 networks
# (trained on ImageNet, won the ILSVRC competition in 2014)
# note: when there is a complete match between your model definition
# and your weight savefile, you can simply call model.load_weights(filename)
assert os.path.exists(weights_path), 'Model weights not found (see "weights_path" variable in script).'
f = h5py.File(weights_path)
for k in range(f.attrs['nb_layers']):
if k >= len(model.layers):
# we don't look at the last (fully-connected) layers in the savefile
break
g = f['layer_{}'.format(k)]
weights = [g['param_{}'.format(p)] for p in range(g.attrs['nb_params'])]
model.layers[k].set_weights(weights)
f.close()
print('Model loaded.')
# build a classifier model to put on top of the convolutional model
top_model = Sequential()
top_model.add(Flatten(input_shape=model.output_shape[1:]))
top_model.add(Dense(256, activation='relu'))
top_model.add(Dropout(0.5))
top_model.add(Dense(5, 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(top_model_weights_path)
# add the model on top of the convolutional base
model.add(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
# compile the model with a SGD/momentum optimizer
# and a very slow learning rate.
model.compile(loss='categorical_crossentropy',
optimizer=optimizers.SGD(lr=1e-4, momentum=0.9),
metrics=['accuracy'])
# prepare data augmentation configuration
train_datagen = ImageDataGenerator(
rescale=1./255,
shear_range=0.2,
zoom_range=0.2,
horizontal_flip=True)
test_datagen = ImageDataGenerator(rescale=1./255)
train_generator = train_datagen.flow_from_directory(
train_data_dir,
target_size=(img_height, img_width),
batch_size=32,
class_mode= 'categorical')
validation_generator = test_datagen.flow_from_directory(
validation_data_dir,
target_size=(img_height, img_width),
batch_size=32,
class_mode= 'categorical')
# fine-tune the model
model.fit_generator(
train_generator,
samples_per_epoch=nb_train_samples,
nb_epoch=nb_epoch,
validation_data=validation_generator,
nb_val_samples=nb_validation_samples)
model.save_weights("fine-tune_weights.h5")
model.save("fine-tune_model.h5", True)
使用
softmax作为输出层的激活函数，它是logistic函数在多class情况下的推广。阅读更多相关信息 here.如果验证误差比训练误差大得多，如您的情况，则表明存在过度拟合。你应该做一些正则化，它被定义为学习算法的任何变化，旨在减少测试错误而不是训练错误。您可以尝试诸如数据增强、提前停止、噪声注入、更积极的 dropout 等方法
如果您有与链接教程中相同的 set-up，请将
train_generator和validation_generator的class_mode更改为categorical并且它将 one-hot 编码你的 classes。