Regression - the output variable takes continuous values.
Example :Given a picture of a person, we have to predict their age on the basis of the given picture
Classification - the output variable takes class labels.
Example: Given a patient with a tumor, we have to predict whether the tumor is malignant or benign.