There are lots of warnings each time run. I wonder what those warning means? Does it means anything I need to change? Thanks, ARNING:theano.configdefaults:install mkl with `conda install mkl-service`: No module named mkl WARNING:theano.tensor.blas:Using NumPy C-API based implementation for BLAS fu…

see Ben’s answer here: [image] Metropolis proposal rejected because location parameter is infinite Modeling Hi! I’m new to Stan, and I’m trying to build a fairly simple model of people’s performances on crosswords. The data is a set of records with a crossword time, which…

Thanks, does it mean the warning does not matter for Metropolis warning before warm-up? Meanwhile, I have other warnings before warm-up, like Warning: left-hand side variable (name=lamba) occurs on right-hand side of assignment, causing inefficient deep copy to avoid aliasing. Also, other warnin…

The warning: Warning: left-hand side variable (name=lamba) occurs on right-hand side of assignment, causing inefficient deep copy to avoid aliasing. can be ignored - it isn’t a problem and the warning has been removed from later versions of the program.

Hi thanks, how about the other like the parameter is nan. I wonder whether is it because the code value underflow or something?

[image] Jenny: How could those parameters be nan? Thanks. One possibility is that the standard deviation for those parameters is zero. If that’s the case, then an nan is benign. Otherwise, you have a serious problem.

Hi, I found my output is like below. It means the sd is 0, so does it matter? Also, I just want to make sure if the Rhat = 1, does it mean that the parameter has converged? Thanks. mean se_mean sd 2.5% 25% 50% 75% 97.5% n_eff Rhat L[1,1,1] 1.0 na…

[image] Jenny: It means the sd is 0, so does it matter? If the sd is 0, then you should be okay. That means that there is no variation in the parameter’s values at all – which is fine if the parameter is supposed to be constrained to a particular value (as is the case for the upper-triangular…

Hi so besides the Rhat as a sign for converge, is there other parameters which can indicate the convergence of the model?

What does this warning means? Should I change the code? And how?

Modeling

mitzimorris November 30, 2018, 10:56pm 2

see Ben’s answer here:

Topic		Replies	Views
Largest Rhat is NA when one parameter is fixed to a constant Modeling specification	27	3518	October 13, 2020
Divergent transitions & BFMI low in a state-space model Modeling	6	1072	August 17, 2017
Warning General	2	1031	July 21, 2017
Two questions: ①Rejecting initial value but still sampling. ②regarding divergent transitions Modeling fitting-issues	11	1166	November 8, 2020
Modelling spatiotemporal random effects results in Modeling ecology , bioinformatics , gaussian-process , hierarchical-model , spatial	21	1818	May 21, 2021

What does this warning means? Should I change the code? And how?

Related topics