Day30 参加职训(机器学习与资料分析工程师培训班)，Tensorflow.keras

今日教学CNN
了解捲积层、池化层、平坦层、丢弃层各层相关係数的设定影响

捲基层: 积层是一组平行的特徵图（feature map），它通过在输入图像上滑动不同的卷积核并执行一定的运算而组成

池化层: 主要是对卷积层的输出进行筛选或者撷取统计上的特徵，并将重要的特徵或资讯保留下来，同时将资料的维度减少

平坦层: 在 CNN 前面几层都是卷积层跟池化层交互转换，后半段会使用多层感知器来稳定判断结果。所以再接入多层感知器前，先必须将矩阵打平成一维的阵列作为输入，然后再串到后面的隐藏层跟输出层。

丢弃层: 提供一个简单且有效率的方式来避免overfitting在训练过程中，每次随机关闭某些神经元，对剩下的神经元进行训练

使用CIFAR10来练习

from tensorflow.keras.datasets import cifar10# Load Data(train_X, train_y), (test_X, test_y) = cifar10.load_data()# Prepare X, y X_train = train_X.astype('float')/255X_test = test_X.astype('float')/255from tensorflow.keras.utils import to_categoricaly_train = to_categorical(train_y, 10)y_test = to_categorical(test_y, 10)#建构CNN网路from tensorflow.keras.models import Sequentialfrom tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, DenseCIFAR10_Model = Sequential()CIFAR10_Model.add(Conv2D(32, (3,3), activation = 'relu', input_shape = (32, 32, 3)))CIFAR10_Model.add(MaxPooling2D((2,2)))CIFAR10_Model.add(Conv2D(64, (3,3), activation = 'relu'))CIFAR10_Model.add(MaxPooling2D((2,2)))CIFAR10_Model.add(Flatten())CIFAR10_Model.add(Dense(128, activation = 'relu'))CIFAR10_Model.add(Dense(10, activation = 'softmax'))CIFAR10_Model.summary()CIFAR10_Model.compile(optimizer = 'adam', loss = 'categorical_crossentropy',                       metrics = ['acc'])from tensorflow.keras.callbacks import TensorBoardCIFAR10_TB = TensorBoard(log_dir = './CIFAR100810', histogram_freq = 1, write_images = True)CIFAR10_Model_History = CIFAR10_Model.fit(X_train, y_train, epochs = 20, batch_size = 256,                                         callbacks = [CIFAR10_TB],                                         validation_split = 0.2)#评估模型Test_loss, Test_acc = CIFAR10_Model.evaluate(X_test, y_test)print(f'Testing loss is {Test_loss} and Testing accuracy is {Test_acc}')