Artificial Intelligence作业代写、CNN留学生作业代做、Python编程设计作业调试、代做Python实验作业代做留学生Processi

School of Computer ScienceThe University of AdelaideArtificial IntelligenceAssignment 2Semester 1, 2019Due 11:55pm, Friday 17th May 2019IntroductionIn this assignment, you will develop classification models to classify noisy input imagesinto two classes: square or circle. Examples are shown in Fig. 1.Figure 1: Samples of noisy images labelled as square (left) and circle (right).Your classification models will use the training and testing sets (thatare available withthis assignment at https://myuni.adelaide.edu.au/courses/45386/assignments)containing many image samples labelled as square or circle. AssignmentYour task is to write Python code which will train and validate the following twoclassification models:1) K-Nearest neighbour (KNN) classifier [30 marks].For the KNN classifier, you can only use standard Python libraries (e.g., numpy)in order to implement all aspects of the training and testing algorithms.Using matplotlib, plot a graph of the evolution of classification accuracy forthe training and testing sets as a function of K, where K = 1 to 10.Clearly identify the value of K for which generalisation is best.Undergraduates can use whatever algorithm they see fit (including exhaustivesearch) and all 30 marks will be available.Post-graduates who implement exhaustive search will be eligible for only 15/30marks. To be eligible for 30/30 you must implement a K-d tree to store andsearch the database.2) Convolutional neural network (CNN) classifier [30 marks].For the convolutional neural network, you should use Tensorflow within JupyterNotebook by modifying the Multilayer Perceptron program supplied with thisassignment. Instructions for installation of Python 3.7, Jupyter and TensorFlow(via a package called miniconda) are in a separate sheet supplied with thisassignment.You should modify the code so that it implements the LeNet CNN structure tothat was presented in lectures. In particular, the LeNet architecture shouldcomprise two convolutional layers (5x5 convolutions), and two hidden fullconnected(dense) layers in addition to the output layer. After each convolutionallayer the architecture should use max pooling to reduce the size by a factor of 2in each axis. After each pooling operation you should use a RELU (RectifiedLinear Unit) activation function. The LeNet will also have three dense layersforming a Multilayer Perceptron (MLP) classifier (you can use the ones already inthe sample implementation. The size of the two hidden-layers in the MLP mustbe 2x and x (where you will need to test different values of x by changing thecode or writing a suitable function). Of course the output layer will have a singleneuron.Undergraduates should experiment training two LeNets, with ‘x’ = 20, ‘x’ = 50and ensure the results are written up carefully, comparing the two networks.Postgraduates should experiment training three LeNets, with ‘x’ = 20, ‘x’ = 50.and ‘x’ = 100 and report the results for all three variants, commenting carefully oneach.Using matplotlib, plot a graph of the evolution of accuracy for the training andtesting sets as a function of the number of epochs, for each of the CNNs youtrain (up to a maximum of 200 epochs).Sample python code that trains and tests a multi-layer perceptron classifier (and can runin a Jupyter Notebook session) is provided with the assignment specification athttps://myuni.adelaide.edu.au/courses/45386/assignments. You should modify this codeto produce your own program.SubmissionYou must submit via MyUni, by the due date, two files:1. A zip file or Jupyter notebook file (.ipynb) containing your code with the twoclassifiers and all implementations described above.2. A pdf file with a short (no more than 2 pages) written report detailing yourimplementation, your results, and an interpretation of the results. The results youshould include are:a. The training and testing accuracies at for KNN, K=1 to 10b. The training and testing accuracies for CNN, x=20, x=50 (and x=100 forpostgrads)This should take the form of a table:Classifier Training Accuracy Testing AccuracyK-NN (k = 1)...K-NN (k = 10)CNN (x = 20)CNN (x = 50)CNN (x = 100)The implementations are worth 30 marks each (see above) and the report is worth 40marks. For full marks your report should include a careful and critical analysis of yourobservations about the performance of the different algorithms and algorithm settings.This assignment is due 11.55pm on Friday 17th May, 2019. If your submissionIs late, the maximum mark you can obtain will be reduced by 25% per day (or partthereof) past the due date or any extension you are granted.本团队核心人员组成主要包括BAT一线工程师，精通德英语！我们主要业务范围是代做编程大作业、课程设计等等。我们的方向领域：window编程 数值算法 AI人工智能 金融统计 计量分析 大数据 网络编程 WEB编程 通讯编程 游戏编程多媒体linux 外挂编程 程序API图像处理 嵌入式/单片机 数据库编程 控制台 进程与线程 网络安全 汇编语言 硬件编程 软件设计 工程标准规等。其中代写编程、代写程序、代写留学生程序作业语言或工具包括但不限于以下范围:C/C++/C#代写Java代写IT代写Python代写辅导编程作业Matlab代写Haskell代写Processing代写Linux环境搭建Rust代写Data Structure Assginment 数据结构代写MIPS代写Machine Learning 作业 代写Oracle/SQL/PostgreSQL/Pig 数据库代写/代做/辅导Web开发、网站开发、网站作业ASP.NET网站开发Finance Insurace Statistics统计、回归、迭代Prolog代写Computer Computational method代做因为专业，所以值得信赖。如有需要，请加QQ：99515681 或邮箱：[email protected] 微信：codehelp QQ：99515681 或邮箱：[email protected] 微信：codehelp