New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Do we need resume_from
now that we have initial_state
?
#2171
Comments
@devmotion @yebai thoughts? |
I don't think we need |
Yes, I think it's redundant and confusing, now that |
I see #2115 did something on this, but it doesn't seem to be "complete", e.g. Lines 100 to 115 in c29d36e
is hit when |
AbstractMCMC@5 has
initial_state
as a mechanism to "resume" a chain.This then begs the question: do we even need
resume_from
that is present in many sampler impls in Turing.jl?Turing.jl/src/mcmc/hmc.jl
Lines 92 to 93 in 39f5d5b
AFAIK the only place
resume_from
is actually overloaded is forChains
:https://github.com/TuringLang/DynamicPPL.jl/blob/ff68206d4b34e230c7e444e91fc60297dd5a5bd0/ext/DynamicPPLMCMCChainsExt.jl#L18-L27
IMO we should replace this with a simple
getstate
or something fromChains
and then just useinitial_state
everywhere to stay consistent with theAbstractMCMC
interface.The text was updated successfully, but these errors were encountered: