I need this to be done in MATLAB. (Not Simulink)

Example An RBF NN is to be traded to approximate the soulinet function y=etsin (3x newrbe 3.4, L in the interval [0, 4]. This is the same task that we examined in Example 3.2 using an MLP NN. To ensure a valid comparison, the interval [0, 4] was again sampled with 21 points separated by 0.2. The MATLAB function trainrbfe was used to train the network. The number of hidden units is 21, and the centers correspond to the input training vectors. Therefore, the system of equations in (3.153) has a unique solution. The MATLAB function trainrbfe uses a Gaussian RBF, and the spread parameter o was set to 0.2. After training, the response of the network to the training patterns was plotted, and is shown in Figure 3.11(a). To test the generalization capability of the network, the same interval was sampled with 401 points, separated by 0.01. This generated the test data set. The network response to this test data set is presented in Figure 3.11(b). The results presented in Example 3.4 represent a case of data overfitting using a RBF NN. In Example 3.2 we saw that overfitting entirely destroys the performance of the MLP NN. Figure 3.11(b) shows that the RBF NN at least maintains acceptable performance even in the case when it overfits the training data. 1 Function approximation 0.8 0.8 Network testing 0.6) 0.6 0.4 0.4 Output: "-", target: "+" 0.2 Output, test set:"-", training set: "+" ht 0.2 . OH -0.2 -0.2 -0.40 0.5 1 1.5 2 2.5 3 3.5 4 -0.45 0.5 1 1.5 2 2.5 3 3.5 4 Input (a) Input (b) Figure 3.11 testing data. RBF neural network response (a) to the training data and (b) to the