Perceptrons:

It is a neural network unit(artificial) that does computations to detect certain features in input data. A perceptron can also be defined as an algorithm used for supervised learning of binary classifiers( decide whether an input belongs to a specific class). In 1943 McCullock and Pitts described a simple brain cell as a logic gate with binary inputs. In 1957 Frank Rosenblatt had an idea that Artificial neurons can stimulate brain principles, which led him to inventing the term PERCEPTRON. The output of a perceptron can only be binary (0 or 1).