DRAM ACMC

[ben18785]

Closes #295

[codecov]

Codecov Report

Merging #982 into master will not change coverage.
The diff coverage is 100%.

@@          Coverage Diff           @@
##           master   #982    +/-   ##
======================================
  Coverage     100%   100%            
======================================
  Files          68     69     +1     
  Lines        7230   7335   +105     
======================================
+ Hits         7230   7335   +105

Impacted Files	Coverage Δ
pints/_mcmc/_dram_ac.py	100% <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 146ea5f...8bd5a27. Read the comment docs.

[MichaelClerx]

Hey @ben18785
Looks like this needs updating to the new structure? Imports a GlobalAdaptiveCovarianceMCMC at some point that's no longer define

[MichaelClerx]

Looks good, but notebook could do with a bit of interpretation of the results.

"Try with 2 proposal kernels instead." Could you point out the issues in the first run in the images? And show how that gets better?

Will make some small suggestions in the code now too :-)

[ben18785]

Looks good, but notebook could do with a bit of interpretation of the results.

"Try with 2 proposal kernels instead." Could you point out the issues in the first run in the images? And show how that gets better?

Will make some small suggestions in the code now too :-)

Very good point -- the notebook was pretty bare before! haha

Have now added more detail and made the notebook a bit more readable. See what you think.

[MichaelClerx]

Thanks Ben, looks great!
To me the traces look wild! Is that a good thing? Massively-mixing mcmc?

[MichaelClerx]

@ben18785 are you happy for me to merge this?

[ben18785]

Yep!

…

On Tue, Mar 31, 2020 at 5:32 PM Michael Clerx ***@***.***> wrote: @ben18785 <https://github.com/ben18785> are you happy for me to merge this? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#982 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABCILKBTJTJFLMLM25DPGNTRKILJNANCNFSM4I33UPCQ> .