Comment on "Estimating model parameters from time series by autosynchronization".