page is served, this fragment will be evaluated and replaced with its value: Tuesday, 15-Jan-2013 19:28:54 EST The decision of when to use SSI, and when to have your page entirely generated by some program [client\ %a] %M% ,\ referer\ %{Referer}i" This would result in error messages such as: [Thu May 12 08:28:57.652118 2011] [core:error] [pid 8777:tid 4326490112] [client ::1:58619] File does not exist: / Use %>s for the final status. %t Time the request was received, in the format [18/Sep/2011:19:18:28 -0400]. The last number indicates the timezone offset from GMT %{format}t The time, in the form given

VM VirtualBox With Oracle Cloud Infrastructure . . . . . 28 1.15.7 Exporting an Appliance to Oracle Cloud Infrastructure . . . . . . . . . . 28 1.15.8 Importing an Instance from Oracle Cloud Infrastructure (2021-10-19) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337 15.2 Version 6.1.26 (2021-07-28) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338 15.3 Version 6.1.24 (2021-07-20) . production environment. See chapter 1.15.7, Exporting an Appliance to Oracle Cloud Infrastructure, page 28. • Guest Additions: shared folders, seamless windows, 3D virtualization. The Oracle VM VirtualBox

sample that uses SOAP::Lite was described in chapter 3.2, Raw web service example for Perl, page 24. 28 4 Using the Main API documentation to write web service clients As “interfaces”, “attributes” and

6: Support Vector Machine Feng Li Shandong University fli@sdu.edu.cn December 28, 2021 Feng Li (SDU) SVM December 28, 2021 1 / 82 Outline 1 SVM: A Primal Form 2 Convex Optimization Review 3 The Kernels 5 Soft-Margin SVM 6 Sequential Minimal Optimization (SMO) Algorithm Feng Li (SDU) SVM December 28, 2021 2 / 82 Hyperplane Separates a n-dimensional space into two half-spaces Defined by an outward 0 means moving it parallely along ω (b < 0 means in opposite direction) Feng Li (SDU) SVM December 28, 2021 3 / 82 Support Vector Machine A hyperplane based linear classifier defined by ω and b Prediction

������的������������ 1 28 �� 1 ������� 2 Collection � Map �� 3 �� 4 Iterator �� 5 � 6 �� 7 ���� API 2 28 ������� 集合��� O 面向存放多个数据的需求 ��用����������� �������������������������� ������用� Java �集合�� 3 28 集合��� 集合��� O 面向存放多个数据的需求 ��用����������� �������������������������� ������用� Java �集合�� 3 28 集合�� 集合�����用���������������合����� �� O 集合类型分类 � Set ��������的������������� ������������的��� �� List ������的��������������� �������的���������������  注意 Java 集合中只能保存引用类型的数据，实际上存放的是对象的引用 而非对象本身。Java API 中的集合类型均定义在 java.util 包中。 4 28 集合�� 集合�����用���������������合����� �� O 集合类型分类 � Set ��������的������������� ������������的��� ��

Lecture 7: K-Means Feng Li Shandong University fli@sdu.edu.cn December 28, 2021 Feng Li (SDU) K-Means December 28, 2021 1 / 46 Outline 1 Clustering 2 K-Means Method 3 K-Means Optimization Problem Problem 4 Kernel K-Means 5 Hierarchical Clustering Feng Li (SDU) K-Means December 28, 2021 2 / 46 Clustering Usually an unsupervised learning problem Given: N unlabeled examples {x1, · · · , xN}; no achieves: High within-cluster similarity Low inter-cluster similarity Feng Li (SDU) K-Means December 28, 2021 3 / 46 Similarity can be Subjective Clustering only looks at similarities, no labels are given

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 645 28 API Reference 647 28.1 Input/Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . the docs. Reducing In [27]: df = DataFrame([[1, np.nan], [1, 4], [5, 6]], columns=[’A’, ’B’]) In [28]: g = df.groupby(’A’) In [29]: g.nth(0) Out[29]: B A 1 NaN 5 6 # this is equivalent to g.first() [’C1’,’C3’]],idx[:,’foo’]] Out[56]: lvl0 a b lvl1 foo foo A0 B0 C1 D0 8 10 D1 12 14 C3 D0 24 26 D1 28 30 B1 C1 D0 40 42 D1 44 46 C3 D0 56 58 ... ... ... A3 B0 C1 D1 204 206 C3 D0 216 218 D1 220

pageAlignHorizontally = true; 25 26 //screen size will be set automatically 27 this.scale.setScreenSize(true); 28 29 //physics system for movement 30 this.game.physics.startSystem(Phaser.Physics.ARCADE); 31 'assets/audio/explosion.ogg'); 25 }, 26 create: function() { 27 this.state.start('MainMenu'); 28 } 29 }; 1 this.preloadBar = this.add.sprite(this.game.world.centerX, this.game.world.centerY if(this.game.input.activePointer.justPressed()) { 26 this.game.state.start('Game'); 27 } 28 } 29 }; 1 this.background = this.game.add.tileSprite(0, 0, this.game.width, this.game.height

AlexNet(8 层)、VGG16(16 层)、 GoogLeNet(22 层)、ResNet50(50 层)、DenseNet121(121 层)等模型相继被提出，同时输入图 片的大小也从28 × 28逐渐增大，变成224 × 224、416 × 416等，这些变化使得网络的总参 数量可达到千万、上亿级别，如图 1.13 所示。 网络规模的增大，使得神经网络的容量也相应增大，从而能够学习到复杂的数据模 10000 张图片作为测试集?test(Test Set)，用来预测或者测试，训练集和测试集共同组成 了整个 MNIST 数据集。 考虑到手写数字图片包含的信息比较简单，每张图片均被缩放到28 × 28的大小，同时 只保留了灰度信息，如图 3.2 所示。这些图片由多人书写，包含了如字体大小、书写风 格、粗细等丰富的样式，使得数据集的分布与真实的手写数字图片的分布尽可能地接近， 从而保证了模型的泛化能力。 1]形状的张量)。图 3.3 演示 了内容为 8 的数字图片的矩阵内容，可以看到，图片中黑色的像素用 0 表示，灰度信息用 0~255 表示，图片中越白的像素点，对应矩阵位置中数值也就越大。 28行28列 图 3.3 图片的表示示意图① ① 素材来自 https://towardsdatascience.com/how-to-teach-a-computer-to-see-wi