1. Train a two-input single perceptron with the following initial weights and data: W = 2 W = -2 bias = 2 Learning Rate: n = 0.1 Data: Pattern 1 2 X1 3 1 x2 1 3 Bias +1 +1 Desired Output -1 +1 (1) Manually update the weights for each pattern in sequential order until the perceptron converges. (2) Report the number of epochs needed for convergence. (3) Plot or sketch how the decision boundary changed after each epoch.