Perceptron learning

  • Select a problem type for generating two classes as 'Linearly separable' or 'Linearly inseparable'.
  • Choose for a number of samples per class and number of iterations the perceptron network must go through.
  • Choose a step size to move over the number of iterations for display of results for the perceptron network.
  • Intialize the perceptron network by clicking on the 'Init perceptron' button.
  • Click on button for display of results according to the given step size.