Jonas Kristoffer Lindeløv

Did you load `lme4` or the like after loading mcp, getting a message like "The following objects are masked from ‘package:mcp fixef ranef’"? It may be trying to use `lme4::ranef()`...

@sjmgarnier I got this to work locally but while it provides much more modeling options (and is easier to maintain/extend), sampling of currently supported models take double the time. mcp...

Cool, `mcp` is henceforth aimed at patient users with powerful computers :-) Yes, for data with < 200 data points, we're talking a slowdown from ~10 secs to ~20 secs...

I'm working on this now and have almost finished making all design decisions. Luckily, I found an implementation that won't negatively affect performance. Expect release in a few months, depending...

@mattmoo I just pushed the development version to branch v0.4 [here](https://github.com/lindeloev/mcp/tree/v0.4). I think you can install it using `remotes::install_github("lindeloev/[email protected]")`. You can fit it and see parameter summaries, including `plot_pars(fit)`. It...

There is good progress on this. I just pushed the latest version to branch v0.4. Install using `remotes::install_github("lindeloev/[email protected]")`. The easiest way to see it in action is running `result =...

I just tested the performance locally on a Ryzen 5 3600. For 55.000 data points and a 15-parameter model with categorical predictors, I get around 1 sample per second. So...

@adrose Unfortunately not. The v0.4 branch currently doesn't run out-of-the-box due to some backwards-incompatible changes in the dependency packages, that I only incorporated into the v0.3 series. I'm presently prioritizing...

Thanks for reporting. Can you post code that reproduces this error? And can you post the stack trace? (use `traceback()` if it isn't returned automatically)

Thanks! This is definitely a bug. The post-processing functions should allow for NA values in the dependent variable and they don't. As a temporary workaround, use a dataset without NA...