https://drive.google.com/file/d/1qy_0mYHk4P--gigHjWgNlbjSSNe24fl2/view?usp=share_link

21. You have been provided with a dataset (wank_dat.m) containing weather information for Ankara from 01/01/1994 to 28/05/1998. The goal of this assignment is to use a Radial Basis Function (RBF) neural network to predict the mean temperature based on nine features: maximum temperature (F), minimum temperature (F), dewpoint (F), precipitation, sea level pressure (in), standard pressure (in), visibility (mi), wind speed (mph), and maximum wind speed (mph). The dataset includes 10 features and 1609 samples. You should use the first 1550 samples for training the network and the last 59 samples for testing its performance. Design an RBF neural network to predict the mean temperature of Ankara based on nine features. Clearly describe the network model, including the number of radial basis functions used, their centers, and the learning rate. Experiment with different network parameters (such as the number of centers and the learning rate) to find the optimal combination that provides the best approximation of the given data. Aim to use the minimum number of hidden units (RBFs) possible while still achieving good performance. Illustrate the effect of using more or less hidden units on the prediction performance. Finally, summarize and conclude the outcomes of this assignment. For this assignment, it is important to clearly explain your methods and results. Make sure to include figures and results to support your explanations. Solutions without sufficient detail and clarity may not receive full marks. It is acceptable to work with your friends on some questions, but you must submit a unique solution for each question. Be sure to carefully follow the instructions provided in the assignment and make sure to address all aspects of the