通过 Keras 2.4.3 和 Tensorflow 2.2 从 SavedModel 中提取特征

Question

我想从我的 CNN 模型的最后一个密集层中提取特征。然而，我对我所做的所有 google 研究感到非常矛盾。 Tensorflow 有很多不同的方法，我正在努力寻找一些有用的方法。

我已经在 CIFAR10 上成功地训练了一个模型。我已将模型保存到一个目录并有一个 saved_model.pb 文件。我已经通过 tensorboard 可视化了模型，但不完全确定我的最后一层的名称。可视化看起来有点混乱。

我怎样才能继续提取这些特征？我想用它们进行 t-SNE 分析。

我正在尝试使用 gfile 加载 pb 图，但不确定这是否正确 approach.Thank 你。

import tensorflow.compat.v1 as tf
tf.disable_v2_behavior()
from tensorflow.python.platform import gfile


pb_graph_file = '../data/processed/saved_models/saved_model.pb'

f = gfile.GFile(pb_graph_file, 'rb')
graph_def = tf.GraphDef()
f.close()

我的 Keras Sequential 模型如下所示：

    """
    This is the CNN model's architecture
    """
    weight_decay = 1e-4
    model = Sequential()
    model.add(Conv2D(32, (3, 3), activation = 'relu', kernel_initializer = 'he_normal', kernel_regularizer = l2(weight_decay), padding = 'same', input_shape = (32, 32, 3)))
    model.add(BatchNormalization())
    model.add(Conv2D(32, (3, 3), activation = 'relu', kernel_initializer = 'he_normal', kernel_regularizer = l2(weight_decay), padding = 'same'))
    model.add(BatchNormalization())
    model.add(MaxPooling2D((2, 2)))
    model.add(Dropout(0.2))

    model.add(Conv2D(64, (3, 3), activation = 'relu', kernel_initializer = 'he_normal', kernel_regularizer = l2(weight_decay), padding='same'))
    model.add(BatchNormalization())
    model.add(Conv2D(64, (3, 3), activation = 'relu', kernel_initializer = 'he_normal', kernel_regularizer = l2(weight_decay), padding='same'))
    model.add(BatchNormalization())
    model.add(MaxPooling2D((2, 2)))
    model.add(Dropout(0.3))

    model.add(Conv2D(128, (3, 3), activation = 'relu', kernel_initializer = 'he_normal', kernel_regularizer = l2(weight_decay), padding='same'))
    model.add(BatchNormalization())
    model.add(Conv2D(128, (3, 3), activation = 'relu', kernel_initializer = 'he_normal', kernel_regularizer = l2(weight_decay), padding='same'))
    model.add(BatchNormalization())
    model.add(MaxPooling2D((2, 2)))
    model.add(Dropout(0.4))

    # model.add(Conv2D(256, (3, 3), activation = 'relu', kernel_initializer = 'he_uniform', kernel_regularizer = l2(weight_decay), padding='same'))
    # model.add(Conv2D(256, (3, 3), activation = 'relu', kernel_initializer = 'he_uniform', kernel_regularizer = l2(weight_decay), padding='same'))
    # model.add(MaxPooling2D((2, 2)))

    model.add(Flatten())
    # model.add(Dense(128, acti vation='relu', kernel_initializer = 'he_normal', kernel_regularizer = l2(weight_decay)))
    # model.add(BatchNormalization())
    # model.add(Dropout(0.5))
    # output layer
    model.add(Dense(10, activation = 'softmax'))

    # optimize and compile model
    opt = Adam(learning_rate = 1e-3)
    model.compile(optimizer = opt, loss = 'categorical_crossentropy', metrics = ['accuracy'])

    return model

Answer 1

首先使用 model.summary().

获取所需图层的名称

然后在下面给定的代码中使用该图层的名称代替 desired_layer：

from keras.models import Model
extractor = Model(inputs=model.inputs, outputs=model.get_layer(desired_layer).output)
features = extractor.predict(x)

此处x是您要从中提取特征的数据。

通过 Keras 2.4.3 和 Tensorflow 2.2 从 SavedModel 中提取特征

Extracting Features from a SavedModel via Keras 2.4.3 and Tensorflow 2.2

python

keras

tensorflow

tensorflow2.0