GitLab

The layout of parameters sections was wrong.
Also pruned some dead code.

* ls

* Changes to ode.rst to make docs appear.

* Should be working now

* add test for xarray input to sample_posterior_predictive
* support for xarray.Dataset as trace argument to sample_posterior_predictive and fast_sample_posterior_predictive.
* closes #3828
Co-authored-by: Michael Osthege <zufallsprinzip@hotmail.de>
Co-authored-by: rpgoldman <rpgoldman@goldman-tribe.org>

* Demonstrate issue 3793.

* Give Data model-relative names.

Use Model.name_for() to relativize pm.Data names.

* Move test description into docstring

* mention changes to pm.Data naming

closes #3793
Co-authored-by: michaelosthege <thecakedev@hotmail.com>

Fix computation of samples argument in sample_posterior_predictive
Solves #3840 

Previously, if the argument to sample_posterior_predictive was a list of points, AND the samples argument was not supplied, then spp would error out, because the code for computing samples assumed that the trace argument was a MultiTrace.

* updated RELEASE-NOTES.md

Added the deprecation of `sd` with `sigma` in newer version with DeprecationWarning on usage of `sd`.

* Update RELEASE-NOTES.md

* partially fixes #3688

* fixes #3688 sd deprecated to sigma

Fixed small but critical typo in Theano Op example (#3832)

* include n_tune, n_draws and t_sampling in SamplerReport
* count tune/draw samples instead of trusting parameters (because of KeyboardInterrupt)
* fall back to tune and len(trace) if tune stat is unavailable
* add test for SamplerReport n_tune and n_draws
* clarify that n_tune are not necessarily in the trace
* use actual number of chains to compute totals
* mention new SamplerReport properties in release notes
Co-authored-by: Michael Osthege <m.osthege@fz-juelich.de>

Second try to vectorize sample_posterior_predictive.

Co-Authored-By: Thomas Wiecki <thomas.wiecki@gmail.com>

I believe that this reshaping -- which flattened the samples -- was a hold-over from the pointwise variant.  It was causing test failures.

Local class declared after reference.

The contextvars library is used to make the vectorized version of sample_posterior_predictive compatible with the legacy version.
contextvars was added in python 3.7, but there is a compatibility library for 3.6
This is an attempt to ensure it will be loaded in python 3.6

When sampling fast_posterior_predictive gets some values with a degenerate extra dimension (e.g., 1 row of 200 values, instead of just 200 values: (1, 200) instead of (200,) ).  Trim these off.
It would be better if I could figure out where these are coming from and avoid that.

More carefully check the results of posterior predictive sampling with pm.Data.

Per @lucianopaz suggestion, I have renamed my new version as fast_sample_posterior_predictive(), and restored the old version.
This is not just nice for testing, but also is critical because we cannot currently vectorize the sampling for multivariate distributions.

Created a new _TraceDict class that represents a trace as a dictionary of variable names to numpy arrays, and use that as the internal representation, per @lucianopaz suggestion.
Made this a class so that it can have varnames and a length, like the MultiTrace.
Added a couple of translation tests.

Incorporate fix from @lucianopaz (98635eb5) and made corresponding changes to posterior_predictive.

Needed to change how theano functions were computed in vectorized variant of spp.

In general, the recursive calls to `draw_values` were not handled properly because I had bugs in the way I handled `ContextManager` objects. I believe those bugs are all fixed now.