使用预训练网络训练的两种方式：Keras Applications、TensorFlow Hub

灰太狼 2023-07-03 03:21 10阅读 0赞

# ![20191009191333910.png][] #

# [日萌社][Link 1] #

[人工智能AI：Keras PyTorch MXNet TensorFlow PaddlePaddle 深度学习实战（不定时更新）][AI_Keras PyTorch MXNet TensorFlow PaddlePaddle]

--------------------

[tensorflow 2.0 画出model网络模型的拓扑图][tensorflow 2.0 _model]

[tensorflow 2.0 的回调函数callbacks（TensorBoard、ModelCheckpoint）][tensorflow 2.0 _callbacks_TensorBoard_ModelCheckpoint]

[TensorBoard视觉化网络每层权重分布、视觉化网络层结构][TensorBoard]

[MSE(均方误差)、MAE(平均绝对误差)][MSE_MAE]

[from\_logits][from_logits]

[CE(Cross-Entropy)、BCE(Binary Cross-Entropy 即Sigmoid+CE)、CCE(Categorical Cross-Entropy 即Softmax+CE)][CE_Cross-Entropy_BCE_Binary Cross-Entropy _Sigmoid_CE_CCE_Categorical Cross-Entropy _Softmax_CE]

[对连续值/离散值进行预处理的两种方式（标准化/one-hot化）、反标准化/逆标准化、字符串预处理][one-hot]

[损失函数loss、指标函数metrics][loss_metrics]

[激活函数 Sigmoid、Tanh、ReLU、Softmax][Sigmoid_Tanh_ReLU_Softmax]

[Batch Normalization][]

[反向传播、梯度下降、学习率、优化器optimizers（GD、SGD、Momentum、AdaGrad、Adam）][optimizers_GD_SGD_Momentum_AdaGrad_Adam]

[权重初始化对于激活函数的选择：随机常态分布/随机正态分布初始化(标准差为0.01或1)、Xavier(Glorot)初始化、He初始化][0.01_1_Xavier_Glorot_He]

[图像增强(IA)、数据预处理][IA]

[混淆矩阵tf.math.confusion\_matrix][tf.math.confusion_matrix]

[使用预训练网络训练的两种方式：Keras Applications、TensorFlow Hub][Keras Applications_TensorFlow Hub]

--------------------

1.使用预训练网络训练的两种方式：Keras Applications、TensorFlow Hub
    2.Keras Applications
    	1.查看Keras Applications模块内置的所有预训练的网络模型
    		>>> import tensorflow
    		>>> help(tensorflow.keras.applications)
    		Help on package tensorflow_core.keras.applications in tensorflow_core.keras:
    		NAME：tensorflow_core.keras.applications - Keras Applications are canned architectures with pre-trained weights.
    		PACKAGE CONTENTS
    			densenet (package)
    			imagenet_utils (package)
    			inception_resnet_v2 (package)
    			inception_v3 (package)
    			mobilenet (package)
    			mobilenet_v2 (package)
    			nasnet (package)
    			resnet (package)
    			resnet50 (package)
    			resnet_v2 (package)
    			vgg16 (package)
    			vgg19 (package)
    			xception (package)
    
    		from tensorflow.keras.applications import densenet
    		from tensorflow.keras.applications import imagenet_utils
    		from tensorflow.keras.applications import inception_resnet_v2
    		from tensorflow.keras.applications import inception_v3
    		from tensorflow.keras.applications import mobilenet
    		from tensorflow.keras.applications import mobilenet_v2
    		from tensorflow.keras.applications import nasnet
    		from tensorflow.keras.applications import resnet
    		from tensorflow.keras.applications import resnet50
    		from tensorflow.keras.applications import resnet_v2
    		from tensorflow.keras.applications import vgg16
    		from tensorflow.keras.applications import vgg19
    		from tensorflow.keras.applications import xception
    
    	2.每个预训练的网络模型中均内置有3个相同名字的重要函数
    		1.preprocess_input：
    			preprocess_input_img = preprocess_input(img)
    			自动对图像数据进行归一化(标准化)。归一化到0到1之间的数值，或者-1到1之间的数值。
    			根据所使用的不同的预训练网络模型，自动对图像数据进行归一化(标准化)到所要求的数值标准之后进行返回。
    			例如：VGG、ResNet-50要求输入图像为0到255的数值，而inception_v3、xception要求输入图像为-1到1的数值。
    
    		2.predict：
    			preds = model.predict(preprocess_input_img)
    			通过传入“preprocess_input预处理(归一化/标准化)过后的”数据进行预测，返回预测的标签信息。
    
    		3.decode_predictions
    			predictions = decode_predictions(preds, top=3)[0]
    			通过传入“predict返回的所预测的”标签信息进行解码，返回具体的类别和概率值。
    			上述代码表示输出预测概率值最高的3个类别信息。
    
    3.TensorFlow Hub
    	1.通过tensorflow-hub官网上列举的所有预训练好的模型进行下载使用
    		tensorflow文档关于tensorflow-hub的介绍(无需翻墙)：https://tensorflow.google.cn/hub?hl=zh_cn
      		tensorflow-hub官网(需翻墙)：https://tfhub.dev/
    	2.比如下面演示下载Inception V3预训练模型为例
    		URL：https://tfhub.dev/google/tf2-preview/inception_v3/classification/4
    		示例代码：
    			m = tf.keras.Sequential([
    				hub.KerasLayer("https://tfhub.dev/google/tf2-preview/inception_v3/classification/4", output_shape=[1001])
    			])
    			m.build([None, 299, 299, 3])  # Batch input shape.
    		Usage信息一栏：一般都会提供训练集对应的标签集下载，还会介绍训练集数据的input size(299x229x3)和归一化数值范围[0,1]。

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 1][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 2][]![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 3][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 4][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 5][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 6][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 7][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 8][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 9][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 10][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 11][]

[20191009191333910.png]: https://img-blog.csdnimg.cn/20191009191333910.png
[Link 1]: http://www.rimengshe.com/
[AI_Keras PyTorch MXNet TensorFlow PaddlePaddle]: https://blog.csdn.net/zimiao552147572/article/details/88867161
[tensorflow 2.0 _model]: https://azusa.blog.csdn.net/article/details/104092884
[tensorflow 2.0 _callbacks_TensorBoard_ModelCheckpoint]: https://azusa.blog.csdn.net/article/details/104100362
[TensorBoard]: https://azusa.blog.csdn.net/article/details/104132982
[MSE_MAE]: https://azusa.blog.csdn.net/article/details/104094483
[from_logits]: https://azusa.blog.csdn.net/article/details/104095436
[CE_Cross-Entropy_BCE_Binary Cross-Entropy _Sigmoid_CE_CCE_Categorical Cross-Entropy _Softmax_CE]: https://azusa.blog.csdn.net/article/details/104095759
[one-hot]: https://azusa.blog.csdn.net/article/details/104099669
[loss_metrics]: https://azusa.blog.csdn.net/article/details/104099800
[Sigmoid_Tanh_ReLU_Softmax]: https://azusa.blog.csdn.net/article/details/104108314
[Batch Normalization]: https://azusa.blog.csdn.net/article/details/104115714
[optimizers_GD_SGD_Momentum_AdaGrad_Adam]: https://azusa.blog.csdn.net/article/details/104106429
[0.01_1_Xavier_Glorot_He]: https://azusa.blog.csdn.net/article/details/104110857
[IA]: https://azusa.blog.csdn.net/article/details/104114416
[tf.math.confusion_matrix]: https://azusa.blog.csdn.net/article/details/104125078
[Keras Applications_TensorFlow Hub]: https://azusa.blog.csdn.net/article/details/104134536
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70]: https://img-blog.csdnimg.cn/20200201172323124.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 1]: https://img-blog.csdnimg.cn/20200201154116550.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 2]: https://img-blog.csdnimg.cn/20200201154121388.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 3]: https://img-blog.csdnimg.cn/20200201154118668.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 4]: https://img-blog.csdnimg.cn/20200201154321310.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 5]: https://img-blog.csdnimg.cn/20200201154317117.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 6]: https://img-blog.csdnimg.cn/20200201154329404.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 7]: https://img-blog.csdnimg.cn/20200201154328844.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 8]: https://img-blog.csdnimg.cn/20200201154330786.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 9]: https://img-blog.csdnimg.cn/20200201154345436.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 10]: https://img-blog.csdnimg.cn/2020020115434035.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 11]: https://img-blog.csdnimg.cn/20200201154352784.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70