loo
loo copied to clipboard
stan-dev
loo R package for approximate leave-one-out cross-validation (LOO-CV) and Pareto smoothed importance sampling (PSIS)
`loo` has been running for four hours now on seven cores. I have no idea when it will finish, so intermittent progress reports would be great (e.g., every 10%). Specifically,...
If the user is providing a function to calculate the log-likelihood contribution of each observation, why does the user also need to pass a call to `relative_eff`? This just seems...
It'd be great if `loo` allowed parallelisation via the [future package](https://github.com/HenrikBengtsson/future) in addition to the existing cores argument. That would make it easy to e.g. use a job scheduler or...
Getting the following error when running `ppc_loo_interval` on a model estimated using CmdStan. For context, I've read the samples into an R stanfit object, and both `y_rep` and `log_lik` were...
In a rstanarm branch, I introduced a `stan_clogit` function which is similar to the `clogit` function in the **survival** package but is actually described in more detail by Stata (...
The package is named as loo, but when psis-loo fails and k-fold-cv is used, then corresponding function could form a object which could be given to compare function.
It looks like there are already tidiers in the **broom** package for rstanarm and stanfit objects, but I didn't see any for loo objects. So I wrote some. I could...
As I understand it, the warning p_waic > 0.4 is there to highlight when the model is over-parameterized, in the sense of approaching one degree of freedom per data point....
Fixes #191 Affects only what is displayed in pareto_k_table. For example, old ``` Computed from 4000 by 262 log-likelihood matrix. Estimate SE elpd_loo -859.1 37.7 p_loo 11.4 2.8 looic 1718.2...