MaggieWang0614

Update to check and output logloss during training; Change total_points name to avoid rewriting N variable (which changes the yhats_train dimension).

I am updating the covariance kernel function in Gaussian process extrapolation to match the common notation of the squared exponential kernel.