This post discusses the design and implementation of a 2-layer MLP neural network for classification. It covers topics such as the architecture of a multilayer perceptron, dataset selection, code implementation, model evaluation, and a comparison with the Scikit Learn library.
Table of contents
Step 1 — Import and Process DataStep 2 — Create Forward PassStep 3— Create Objective Function (Cross Entropy)Step 4— Initialize Weights and BiasesStep 5—Make PredictionsStep 6 — Select a Threshold and Convert Predictions to Classes23 Impressions