Update GP-Latent.ipynb to v4

[bwengals]

Update GP-Latent.ipynb to V4

This notebook was previously merged (see discussion in #371) but was overwritten. This PR re-adds it.

[review-notebook-app[bot]]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[review-notebook-app[bot]]

View / edit / reply to this conversation on ReviewNB

OriolAbril commented on 2022-08-09T04:13:12Z ----------------------------------------------------------------

Line #8.    f_post = idata.posterior["f"].stack(samples=["chain", "draw"]).T

If you use the current arviz development version you can do

f_post = az.extract(idata, var_names="f").transpose("sample", ...)

I assume plot_pg_dist wants the sample dimension to be the first one.

bwengals commented on 2022-09-28T19:14:32Z ----------------------------------------------------------------

Yup it does. plot_gp_dist really needs to be updated to work with InferenceData objects. Really like az.extract, though I have to brag that I'm pretty fast now at typing idata.posterior.thing.stack(sample=['chain', 'draw'])

[review-notebook-app[bot]]

View / edit / reply to this conversation on ReviewNB

OriolAbril commented on 2022-08-09T04:13:13Z ----------------------------------------------------------------

can you add a watermark title right above the watermark code cell? and the -p aesara,aeppl,xarray flag. Ref: https://www.pymc.io/projects/docs/en/stable/contributing/jupyter_style.html

bwengals commented on 2022-09-28T19:13:09Z ----------------------------------------------------------------

Ah yup, added

[bwengals]

Ah yup, added

---

View entire conversation on ReviewNB

[bwengals]

Yup it does. plot_gp_dist really needs to be updated to work with InferenceData objects. Really like az.extract, though I have to brag that I'm pretty fast now at typing idata.posterior.thing.stack(sample=['chain', 'draw'])

---

View entire conversation on ReviewNB