in particular, the priors have a gazzillion points that take forever to read in. If we supply a --downsample option, then we should be able to decimate the number of points that need to be read in before we go off and try to read them (ie: after utils.load but before the loop that calculates quantiles for the process)
in particular, the priors have a gazzillion points that take forever to read in. If we supply a --downsample option, then we should be able to decimate the number of points that need to be read in before we go off and try to read them (ie: after utils.load but before the loop that calculates quantiles for the process)