search

probcomp / GenExperimental.jl

Featherweight embedded probabilistic programming language and compositional inference programming library
MIT License
17 stars 2 forks source link

Parallelize across model samples when training amortized inference networks #8

Closed marcoct closed 7 years ago

marcoct commented 7 years ago

This could be useful on e.g. probcomp-?

marcoct commented 7 years ago

I did this for the goal inference demo. You parallelize across independent minibatch samples. Seems to be useful.