Generate 2000 points from two equally weighted spherical Gaussians N(0, I), N((3,0, . . . ,0), I) in R^p, p= 1,11,21, . . .101 (note, you have to first flip a coin to decide from which Gaussian to sample), where I is the identity matrix and the centers of Gaussians are distance 3 apart. Implement 1-NN and 3-NN classifiers. Test the resulting classifier on aseparately generated dataset with 1000 pts. Plot the error rate as a function of p. Observations?