Repository navigation
Open
Labels
Description
Hi Ben,
I tried the 500th example of stan_demo, but it seems not to be working. I ran the code from RStudio cloud. Here are my codes:
library(RStan)
stan_demo()
500
Here are the console log:
> K <- 8
> D <- 4
> N <- 500
> y <-
>+ structure(c(1, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0,
>+ 1, 0, 1, 1, 0, 0, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 0, 1, 0, .... [TRUNCATED]
> x <-
+ structure(c(-0.56947710721168, 0.0686981186395276, 0.807151998639757,
+ 0.311832939914023, 0.670993954113173, -0.914759441112067, -0.1375718 .... [TRUNCATED]
SAMPLING FOR MODEL 'probit-multi-good' NOW (CHAIN 1).
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Rejecting initial value:
Error evaluating the log probability at the initial value.
Exception: Phi: x is nan, but must not be nan! (in 'model4f66680104f2_probit_multi_good' at line 26)
Rejecting initial value:
Log probability evaluates to log(0), i.e. negative infinity.
Stan can't start sampling from this initial value.
Initialization between (-2, 2) failed after 100 attempts.
Try specifying initial values, reducing ranges of constrained values, or reparameterizing the model.
[1] "Error in sampler$call_sampler(args_list[[i]]) : Initialization failed."
error occurred during calling the sampler; sampling not done
Activity
bob-carpenter commentedon Nov 30, 2018
Thanks. The set of examples was just scraped and many of them don't work. In my opinion, it should be edited down to a small set of working examples. The decision's up to @bgoodri.
bob-carpenter commentedon Nov 30, 2018
Also, this is partly an RStan issue for just scraping this repo and partly this repo's fault for not having consistent working examples. So there are two ways to fix it---either by pruning example-models or by cleaning up RStan. Cleaning the example-models repo is probably best.
bgoodri commentedon Nov 30, 2018
bgoodri commentedon Nov 30, 2018
In this case, I think it is fine if you specify
init_r = 1or thereabouts. But we don't have a way of knowing that instan_demo. I guess the example-models repo could optionally specify initial values.