Ven Popov

icon indicating a website link http://venpopov.com

icon company University of Zurich icon location Zurich, Switzerland icon location Senior Scientist in Computation Modeling

Results 77 comments of Ven Popov

Adds feature to make inits from fit and draws objects

just a comment on the windows tests, not related to this PR per se - we are seeing the same failures for other packages, so definitely unrelated to the PR....

Adds feature to make inits from fit and draws objects

just a comment on the windows tests, not related to this PR per se - we are seeing the same failures for other packages, so definitely unrelated to the PR....

"numeric overflow" error for models with von Mises distributions and large kappa values

Fixed by #3036

knowing the initial values for sampler even I don't set up the initials

Hi folks, any updated on this? I have a model that runs fine 9 out of 10 times on the same data, but occasionally one chain would just go haywire,...

fixing mu parameter via mu = some_value?

Didn't realize the order, I guess it was newer because I never knew it was possible that way. I like it because it is clearer and shorter. Terms with ~...

(potentially) Incorrect output of pvon_mises()

Hmm, stan produces the same results as brms. I guess both brms and stan parametrize the cdf as $P((Y+\pi\) mod 2\pi - \pi)-P(-\pi)$ regardless of where mu is. But that...

custom families and `mu` parameter

I see that in the description for `brmsformula` you have the following paragraph: > Parameter mu exists for every family and can be used as an alternative to specifying terms...

custom families and `mu` parameter

Definitely not a high priority. Perhaps for v3.0.0 :) I was thinking about this, because we are building [a package for cognitive modeling](https://github.com/venpopov/bmm) that uses `brms` as an interface to...

custom families and `mu` parameter

closing this to consolidate all related issues in #1660

Add special case for derivative of modified_bessel_function(0,x) to greatly improve model estimation speed

@andrjohns I can try to implement this after our discussion in the other issue. Do you think that the conditional approach checking if the order of the bessel function is...