Re-run putting workflow

[MarcoGorelli]

use inferencedata and arviz-darkgrid

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

MarcoGorelli commented on 2020-12-11T13:47:45Z
----------------------------------------------------------------

oops, this is wrong, fixing it now

[codecov]

Codecov Report

Merging #4325 (4f4076c) into master (2a38198) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master    #4325   +/-   ##
=======================================
  Coverage   87.54%   87.54%           
=======================================
  Files          88       88           
  Lines       14272    14272           
=======================================
  Hits        12495    12495           
  Misses       1777     1777           

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

michaelosthege commented on 2020-12-12T20:46:08Z
----------------------------------------------------------------

Instead of plotting 50 posterior samples, you could plot a density with

pst = logit_trace.posterior.stack(sample=("chain", "draw"))
pm.gp.util.plot_gp_dist(
    ax=ax, 
    x=t,
    samples=(pst["a"] * t + pst["b"]).values.T
)

[twiecki]

/pre-commit-run

[MarcoGorelli]

/pre-commit-run

😄 that would require #4330 to be in first

Also, this PR passes the style checks so it wouldn't do anything anyway

[twiecki]

/pre-commit-run

[twiecki]

Yeah, just checking. It wouldn't comment or anything?

[MarcoGorelli]

Not currently, no, it just puts the 'rocket' and 'eyes' emojis (see above). You can see that it was triggered in the 'actions' tab https://github.com/pymc-devs/pymc3/actions

[twiecki]

Very cool!

[MarcoGorelli]

closing as this will go in pymc-examples