Optimization and of Parameters for Complex Data: Experimental Project and Design Exercise

Objectives:  Generate scripts to model and train a Multi-Layer Perceptron and RBF networks using Matlab built in functions.  Utilize an iterative approach to train models with pre-selected and generative parameters  Train multiple instances of selected network models, and preserve model parameters for future use.  Rank model performance using Confusion Matrix, ROC and performance metrics, and provide justification for selected model performance. Requirements/Task(s): Setup: You’ll need to access the websites for each data set and look at the statistics for the vector components in each sample. The goal of the two models we’ll be aiming for will be to construct optimally scaled designs that generate reliable performance. and there are more Requirements at the file