This doesn't work because the correct value for a sub-sampled batch will be smaller than for a full data set.
On Fri, Dec 16, 2011 at 10:05 AM, Dmitriy Lyubimov <[email protected]>wrote: > if it > makes sense to find a better guess for lambda by just doing an R > simulation on a randomly subsampled data before putting it into > pipeline? or there's a fundamental problem with this approach? >
