3 ms·
Yes, this is basically a distributed version of the issues discussed in psychology regarding researcher degrees of freedom with a single data set. If you throw
by jdmichal 10y ago
Yes, this is basically a distributed version of the issues discussed in psychology regarding researcher degrees of freedom with a single data set. If you throw enough models at the data, one of them is bound to stick, whether it is predictive or not.