Merge branch 'master' into plot_bpv

.azure-pipelines/azure-pipelines-external.yml

@@ -43,6 +43,8 @@ jobs:
     displayName: 'Debug information'
 
   - script: |
+      sudo apt-get update
+      sudo apt-get install jags
       python -m pip install --upgrade pip
 
       if [ "$(pytorch.version)" = "latest" ]; then

.azure-pipelines/azure-pipelines-wheel.yml

@@ -34,7 +34,8 @@ jobs:
   - script: |
       python -m pip install --upgrade pip
       python -m pip install --no-cache-dir -r requirements.txt
-      pip install wheel
+      python -m pip install wheel
+      python -m pip install twine
     displayName: 'Install requirements'
 
   - script: |
@@ -43,9 +44,10 @@ jobs:
     displayName: 'Build a wheel'
 
   - script: |
-      cd dist
-      ls -lh
-      ls | grep *.whl | xargs python -m pip install
+      mkdir install_test
+      cd install_test
+      ls -lh ../dist
+      python -m pip install ../dist/*.whl
       python -c "import arviz as az; print(az);print(az.summary(az.load_arviz_data('non_centered_eight')))"
       cd ..
     displayName: 'Install and test the wheel'
@@ -60,3 +62,7 @@ jobs:
       pathtoPublish: 'dist'
       artifactName: 'arviz_wheel_dist'
     displayName: 'Publish the wheel'
+
+  - script: |
+      python -m twine upload -u __token__ -p $(PYPI_PASSWORD) --skip-existing dist/*
+    displayName: 'Upload wheel to PyPI'

CHANGELOG.md

@@ -3,16 +3,38 @@
 ## v0.x.x Unreleased
 
 ### New features
-* loo-pit plot. The kde is computed over the data interval (this could be shorter than [0, 1]). The hdi is computed analitically (#1215) 
+* loo-pit plot. The kde is computed over the data interval (this could be shorter than [0, 1]). The HDI is computed analitically (#1215)
+* Added `html_repr` of InferenceData objects for jupyter notebooks. (#1217)
+* Added support for PyJAGS via the function `from_pyjags`. (#1219 and #1245)
+* `from_pymc3` can now retrieve `coords` and `dims` from model context (#1228, #1240 and #1249)
+* `plot_trace` now supports multiple aesthetics to identify chain and variable
+  shape and support matplotlib aliases (#1253)
+* `plot_hdi` can now take already computed HDI values (#1241)
 
 ### Maintenance and fixes
-
+* Include data from `MultiObservedRV` to `observed_data` when using
+  `from_pymc3` (#1098)
 * Added a note on `plot_pair` when trying to use `plot_kde` on `InferenceData`
   objects. (#1218)
+* Added `log_likelihood` argument to `from_pyro` and a warning if log likelihood cannot be obtained (#1227)
+* Skip tests on matplotlib animations if ffmpeg is not installed (#1227)
+* Fix hpd bug where arguments were being ignored (#1236)
+* Remove false positive warning in `plot_hdi` and fixed matplotlib axes generation (#1241)
+* Change the default `zorder` of scatter points from `0` to `0.6` in `plot_pair` (#1246)
+* Update `get_bins` for numpy 1.19 compatibility (#1256)
+* Fixes to `rug`, `divergences` arguments in `plot_trace` (#1253)
 
 ### Deprecation
+* Using `from_pymc3` without a model context available now raises a
+  `FutureWarning` and will be deprecated in a future version (#1227)
+* In `plot_trace`, `chain_prop` and `compact_prop` as tuples will now raise a
+  `FutureWarning` (#1253)
+* `hdi` with 2d data raises a FutureWarning (#1241)
 
 ### Documentation
+* A section has been added to the documentation at InferenceDataCookbook.ipynb illustrating the use of ArviZ in conjunction with PyJAGS. (#1219 and #1245)
+* Fixed inconsistent capitalization in `plot_hdi` docstring (#1221)
+* Fixed and extended `InferenceData.map` docs (#1255)
 
 ## v0.8.3 (2020 May 28)
 ### Maintenance and fixes
@@ -281,4 +303,3 @@
 ## v0.3.0 (2018 Dec 14)
 
 * First Beta Release
-

GOVERNANCE.md

@@ -93,8 +93,12 @@ Council Members will have the responsibility of
 * Make decisions when regular community discussion doesn’t produce consensus on an issue in a reasonable time frame.
 * Make decisions about strategic collaborations with other organizations or individuals.
 * Make decisions about the overall scope, vision and direction of the project.
+* Developing funding sources
+* Deciding how to disburse funds with consultation from Core Contributors
 
-Note that each individual council member does not have the power to unilaterally wield these responsibilities, but the council as a whole must jointly make these decisions. In other words, Council Members are first and foremost Core Contributors, but only when needed they can collectively make decisions for the health of the project.
+The council may choose to delegate these responsibilities to sub-committees. If so, Council members must update this document to make the delegation clear.
+
+Note that individual council member does not have the power to unilaterally wield these responsibilities, but the council as a whole must jointly make these decisions. In other words, Council Members are first and foremost Core Contributors, but only when needed they can collectively make decisions for the health of the project.
 
 ArviZ will be holding its first election to determine its initial council in the coming weeks and this document will be updated.
 
@@ -182,9 +186,9 @@ Each voter can vote zero or more times, once per each candidate. As this is not
 
 #### Voting Criteria For Future Elections
 Voting for first election is restricted to establish stable governance, and to defer major decision to elected leaders
-* For the first election only the folks in Slack can vote (excluding GSOC students)
+* For the first election only the people registered following the guidelines in elections/ArviZ_2020.md can vote
 * In the first year, the council must determine voting eligibility for future elections between two criteria:
-  * Those with commit bits
+  * Core contributors
   * The contributing community at large
 
 ### Core Contributors
@@ -197,9 +201,18 @@ Current Core Contributors can nominate candidates for consideration by the counc
 can make the determination for acceptance with a process of their choosing.
 
 #### Current Core Contributors
-* Will be updated with Core Contributor list during first election
+* Oriol Abril-Pla (@OriolAbril) 
+* Alex Andorra (@AlexAndorra)
+* Seth Axen (@sethaxen)
+* Colin Carroll (@ColCarroll)
+* Robert P. Goldman (@rpgoldman)
+* Ari Hartikainen (@ahartikainen)
+* Ravin Kumar (@canyon289)
+* Osvaldo Martin (@aloctavodia)
+* Mitzi Morris (@mitzimorris)
+* Du Phan (@fehiepsi)
+* Aki Vehtari (@avehtari)
 
 #### Core Contributor Responsibilities
 * Enforce code of conduct
 * Maintain a check against Council
-

arviz/data/__init__.py

@@ -7,6 +7,7 @@
 from .io_cmdstan import from_cmdstan
 from .io_cmdstanpy import from_cmdstanpy
 from .io_dict import from_dict
+from .io_pyjags import from_pyjags
 from .io_pymc3 import from_pymc3, from_pymc3_predictions
 from .io_pystan import from_pystan
 from .io_emcee import from_emcee
@@ -24,6 +25,7 @@
     "dict_to_dataset",
     "convert_to_dataset",
     "convert_to_inference_data",
+    "from_pyjags",
     "from_pymc3",
     "from_pymc3_predictions",
     "from_pystan",

arviz/data/inference_data.py

@@ -3,13 +3,16 @@
 from collections.abc import Sequence
 from copy import copy as ccopy, deepcopy
 from datetime import datetime
+from html import escape
 import warnings
+import uuid
 
 import netCDF4 as nc
 import numpy as np
 import xarray as xr
+from xarray.core.options import OPTIONS
 
-from ..utils import _subset_list
+from ..utils import _subset_list, HtmlTemplate
 from ..rcparams import rcParams
 
 SUPPORTED_GROUPS = [
@@ -125,14 +128,39 @@ def __init__(self, **kwargs):
                     self._groups_warmup.append(key)
 
     def __repr__(self):
-        """Make string representation of object."""
+        """Make string representation of InferenceData object."""
         msg = "Inference data with groups:\n\t> {options}".format(
             options="\n\t> ".join(self._groups)
         )
         if self._groups_warmup:
             msg += "\n\nWarmup iterations saved ({}*).".format(WARMUP_TAG)
         return msg
 
+    def _repr_html_(self):
+        """Make html representation of InferenceData object."""
+        display_style = OPTIONS["display_style"]
+        if display_style == "text":
+            html_repr = f"<pre>{escape(repr(self))}</pre>"
+        else:
+            elements = "".join(
+                [
+                    HtmlTemplate.element_template.format(
+                        group_id=group + str(uuid.uuid4()),
+                        group=group,
+                        xr_data=getattr(  # pylint: disable=protected-access
+                            self, group
+                        )._repr_html_(),
+                    )
+                    for group in self._groups_all
+                ]
+            )
+            formatted_html_template = HtmlTemplate.html_template.format(  # pylint: disable=possibly-unused-variable
+                elements
+            )
+            css_template = HtmlTemplate.css_template  # pylint: disable=possibly-unused-variable
+            html_repr = "%(formatted_html_template)s%(css_template)s" % locals()
+        return html_repr
+
     def __delattr__(self, group):
         """Delete a group from the InferenceData object."""
         if group in self._groups:
@@ -346,20 +374,28 @@ def _group_names(self, groups, filter_groups=None):
     def map(self, fun, groups=None, filter_groups=None, inplace=False, args=None, **kwargs):
         """Apply a function to multiple groups.
 
+        Applies ``fun`` groupwise to the selected ``InferenceData`` groups and overwrites the
+        group with the result of the function.
+
         Parameters
         ----------
-        fun: callable
-            Function to be applied to each group.
-        groups: str or list of str, optional
+        fun : callable
+            Function to be applied to each group. Assumes the function is called as
+            ``fun(dataset, *args, **kwargs)``.
+        groups : str or list of str, optional
             Groups where the selection is to be applied. Can either be group names
             or metagroup names.
-        inplace: bool, optional
+        filter_groups : {None, "like", "regex"}, optional
+            If `None` (default), interpret var_names as the real variables names. If "like",
+            interpret var_names as substrings of the real variables names. If "regex",
+            interpret var_names as regular expressions on the real variables names. A la
+            `pandas.filter`.
+        inplace : bool, optional
             If ``True``, modify the InferenceData object inplace,
             otherwise, return the modified copy.
-        args: array_like, optional
-            Positional arguments passed to ``fun``. Assumes the function is called as
-            ``fun(dataset, *args, **kwargs)``.
-        **kwargs: mapping, optional
+        args : array_like, optional
+            Positional arguments passed to ``fun``.
+        **kwargs : mapping, optional
             Keyword arguments passed to ``fun``.
 
         Returns
@@ -376,10 +412,40 @@ def map(self, fun, groups=None, filter_groups=None, inplace=False, args=None, **
 
             In [1]: import arviz as az
                ...: idata = az.load_arviz_data("non_centered_eight")
-               ...: idata_shifted_obs = idata.map(lambda x: x + 3, groups="observed_RVs")
+               ...: idata_shifted_obs = idata.map(lambda x: x + 3, groups="observed_vars")
                ...: print(idata_shifted_obs.observed_data)
                ...: print(idata_shifted_obs.posterior_predictive)
 
+        Rename and update the coordinate values in both posterior and prior groups.
+
+        .. ipython::
+
+            In [1]: idata = az.load_arviz_data("radon")
+               ...: idata = idata.map(
+               ...:     lambda ds: ds.rename({"gamma_dim_0": "uranium_coefs"}).assign(
+               ...:         uranium_coefs=["intercept", "u_slope", "xbar_slope"]
+               ...:     ),
+               ...:     groups=["posterior", "prior"]
+               ...: )
+               ...: idata.posterior
+
+        Add extra coordinates to all groups containing observed variables
+
+        .. ipython::
+
+            In [1]: idata = az.load_arviz_data("rugby")
+               ...: home_team, away_team = np.array([
+               ...:     m.split() for m in idata.observed_data.match.values
+               ...: ]).T
+               ...: idata = idata.map(
+               ...:     lambda ds, **kwargs: ds.assign_coords(**kwargs),
+               ...:     groups="observed_vars",
+               ...:     home_team=("match", home_team),
+               ...:     away_team=("match", away_team),
+               ...: )
+               ...: print(idata.posterior_predictive)
+               ...: print(idata.observed_data)
+
         """
         if args is None:
             args = []
@@ -427,7 +493,7 @@ def _wrap_xarray_method(
 
             In [1]: import arviz as az
                ...: idata = az.load_arviz_data("non_centered_eight")
-               ...: idata_means = idata._wrap_xarray_method("mean", groups="latent_RVs")
+               ...: idata_means = idata._wrap_xarray_method("mean", groups="latent_vars")
                ...: print(idata_means.posterior)
                ...: print(idata_means.observed_data)