# Can someone explain - Warning: The current proposal will be rejected due to numerical error(s)

**URL:** <https://discourse.julialang.org/t/can-someone-explain-warning-the-current-proposal-will-be-rejected-due-to-numerical-error-s/103409>\
**Category:** New to Julia\
**Tags:** question, turing\
**Created:** [August 31, 2023, 3:11pm UTC](https://discourse.julialang.org/t/can-someone-explain-warning-the-current-proposal-will-be-rejected-due-to-numerical-error-s/103409 "2023-08-31T15:11:50Z")\
**Posts on this page:** 1\
**Showing post:** 2

<div class="post-metadata">

**Author:** ![mbauman](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/mbauman/32/31082_2.png) [@mbauman](https://discourse.julialang.org/u/mbauman)\
**Post date:** [August 31, 2023, 4:41pm UTC](https://discourse.julialang.org/t/can-someone-explain-warning-the-current-proposal-will-be-rejected-due-to-numerical-error-s/103409/2 "2023-08-31T16:41:38Z")

</div>

As I understand, it’s really model- and even workflow-dependent, which makes this hard to diagnose or even describe in general. There’s been lots of questions about this here before ([1](https://discourse.julialang.org/t/turing-jl-warning-the-current-proposal-will-be-rejected-due-to-numerical-error-s/72396), [2](https://discourse.julialang.org/t/turing-jl-nuts-gets-stuck-in-the-current-proposal-will-be-rejected-isfinite-r-true-true-false-true/81428/), [3](https://discourse.julialang.org/t/turing-jl-warning-the-current-proposal-will-be-rejected-due-to-numerical-error-s-isfinite-r-true-false-false-false/93472), [4](https://discourse.julialang.org/t/help-speeding-up-and-reducing-numerical-errors-for-simple-epidemic-model-with-turing-jl/46614/), [5](https://discourse.julialang.org/t/numerical-errors-in-logit-normal-model-using-turing-jl/30602/), …), but most dig into specific models. Perhaps the best general discussion is in [this GitHub issue](https://github.com/TuringLang/Turing.jl/issues/1621#issuecomment-844621714):

> The errors can be real and indicate a problem with the model and/or the sampler settings but they can be ignored in the initial phase when the step size is tuned - depending on the model and the initialization, it can happen that a too large step size results in a non-finite gradient of the log density.

---

_[View the full topic](https://discourse.julialang.org/t/can-someone-explain-warning-the-current-proposal-will-be-rejected-due-to-numerical-error-s/103409)._