DON'T USE "CHATGPT and TENSORFLOW". Please solve this question by using MATLAB:

Questions (100 points): In this assignment, you will train feedforward neural networks with three layers to classify motion data. You can download "Human Activity Recognition Using Smartphones" data set from https://archive.ics.uci.edu/dataset/240/human+activity+recognition+using+smartphones link. Then, unzip UCI HAR Dataset.zip in human+activity+recognition+using+smartphones/ directory into the same folder. In total, there are 10299 movement patterns with 561 features belonging to 6 classes in the data set. These features are explained in detail at human+activity+recognition+using+smartphones/ UCI HAR Dataset/features.txt file. This data set is divided into two as training and test subsets. Training and test sets include 7352 and 2947 pattems, respectively. You can benefit from "Human activity recognition with different artificial neural network based classifiers" named conference paper in the "https://ieeexplore.ieee.org/abstract/document/7960559" link to understand data set and assignment better. - As a first step, load training and test data sets with similar codes to the following. X_train = load("humantactivity+recognition+usingtamartphones/UCI HAR Dataset/train/X_train.txt ); Y_train = load("human+activity+recognition+usingtsmartphones/UCI HAR Dataset/train/Y_train.txt"); X_test = load("humantactivity+recognition+usingtsmartphones/UCI HAR Dataset/test/X_test_txt"); Y_test = load("humantactivity+recognitiontusingtamartphones/UCI HAR Dataset/test/Y_test.txt"); - Here, \( Y_{\text {_train }} \) and \( Y_{\text {_test }} \) variables include class labels from 1 to 6. You must transform them into the one hot encoding scheme. - After that, you must generate a three-layer neural network, which takes 561 features to the first hidden layer consisting of 200 neurons with hyperbolic tangent activation functions. The second hidden layer of the neural network consists of 100 neurons with rectified linear unit activation functions. The output layer consists of 6 neurons with either sigmoid or softmax activation functions. Finally, cost function will be chosen either MSE or Cross Entropy. To do these, you benefit from following codes. h1_zize =200; h2_zize =100; epochlinit =80; input_size =size(Xtrain, 2); Whle =(rand(h1 _size, inpurt_size + 1) -0.5)/100; Wh 2e=(rand(h2_size,h1_size+1)0.5)/100; Woe =( rand ( length ( unique \( \left.\left(Y_{\text {_test }} ight) ight) \), h2_size + 1)0.5)/100; - Moreover, the following cross-entropy (CE) cost function definition should be utilized when requested. JCE=h1l=1hi=1qdi(l)ln(oi(l)) Here, q is the number of output neurons, h is the number of data set patterns, di and oi are the desired and actual output values for ith neuron at lth pattern, respectively. - In addition, the following mean squared error cost (MSE) function definition should be utilized when requested. JMSE=2h1i=1hi=1q(di(l)oi(l))2 Here, q is the number of output neurons, h is the number of data set patterns, di and oi are the desired and actual output values for ith neuron at lth pattern, respectively. 1. Perform a training using sigmoid activation function on output layer and mean squared error metric with learning rate eta =0.001 for 80 epochs on training set data. Using the plot function, show the change of MSE loss over training and test sets in the same figure with respect to epoch number ( 20 points). Find the occurred minimum loss value and its epoch number on each set (4 points). Generate a confusion matrix for training and test sets using weights at the end of the training (16 points). Calculate the ratio of successful classifications to all pattems inside each set using similar codes to the following ( 6 points). All related codes should be explained to take these points. trainingSuccessRate=sum(diag(confusionMatrixTraining))/sum(sum(confusionMatrixTraining))testSuccessate=sum(diag(confusionMatrixTest))/sun(sum(confusionMatrixTest)) 2. Repeat first question using softmax activation function on output layer and cross entropy cost function instead of sigmoid and MSE (46 points). 3. Comment on difference between sigmoid +MSE and softmax +CE utilization considering their advantages and drawbacks ( 8 points)