Illustration of perceptron learning for two-dimensional input

Consider two linearly separable classes, where each class consists of two-dimensional input pattern vectors. An example of two-class classification problem is shown below. The input vectors belonging to each class are shown in different colours in the following figure. The line is defined by the equation w1x+w2y=θ. The straight line is defined by the initial values of the weights and the threshold.

After the convergence of weights, the line separates the two classes. This is shown in the following figure.