Skip to content
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

use Turing through imports #123

Closed
wants to merge 6 commits into from
Closed

use Turing through imports #123

wants to merge 6 commits into from

Conversation

ChrisRackauckas
Copy link
Member

No description provided.

@codecov
Copy link

codecov bot commented Jan 6, 2020

Codecov Report

Merging #123 into master will not change coverage.
The diff coverage is 100%.

Impacted file tree graph

@@           Coverage Diff           @@
##           master     #123   +/-   ##
=======================================
  Coverage   91.17%   91.17%           
=======================================
  Files           7        7           
  Lines         238      238           
=======================================
  Hits          217      217           
  Misses         21       21
Impacted Files Coverage Δ
src/DiffEqBayes.jl 83.33% <ø> (ø) ⬆️
src/turing_inference.jl 94.59% <100%> (ø) ⬆️

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 9ea529f...a575a88. Read the comment docs.

@Vaibhavdixit02
Copy link
Member

Will require new release of AdvancedHMC TuringLang/AdvancedHMC.jl#160, once it's in this should be good to go. Works for me locally.

@xukai92
Copy link
Contributor

xukai92 commented Jan 8, 2020

The new release is there.

@Vaibhavdixit02
Copy link
Member

Thanks @xukai92!

@ChrisRackauckas
Copy link
Member Author

@xukai92 can we get some help figuring out what Turing is trying to iterate?

@xukai92
Copy link
Contributor

xukai92 commented Jan 11, 2020

Do you mean why we do this DynamicPPL thing?

@Vaibhavdixit02
Copy link
Member

@ChrisRackauckas that happens because https://github.com/TuringLang/Turing.jl/blob/master/src/inference/Inference.jl#L856's types, the MvNormal is <: both Distribution and MultivariateDistribution, but the MultivariateDistribution method does not accept AbstractArray type, so when you don't matrixize in the dot_tilde call here, it can't iterate. On changing it though, I am encountering another error which I can't understand https://travis-ci.org/JuliaDiffEq/DiffEqBayes.jl/jobs/635608371?utm_medium=notification&utm_source=github_status

@xukai92
Copy link
Contributor

xukai92 commented Jan 13, 2020

It's might be related to what @mohamed82008 is trying to fix.

@mohamed82008
Copy link
Contributor

Yes that looks related to TuringLang/Turing.jl#1043. I will take a look next week. Please open an issue in Turing to track it, happy to investigate when I get some time.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants