[DOC] Update documentation

Author: xuyxu

CHANGELOG.rst

@@ -4,26 +4,29 @@ Changelog
 [Beta]
 ------
 
-Added
-~~~~~
-* [ENH] Add set_scheduler | @xuyxu
-* [ENH] Add AdversarialTrainingClassifier and AdversarialTrainingRegressor | @xuyxu
-* [ENH] Add SnapshotEnsembleClassifier and SnapshotEnsembleRegressor | @zzzzwj and @xuyxu
-* [ENH] Add model validation and serialization | @ghost-ronin and @xuyxu
-* [MNT] Add CI and maintenance tools | @xuyxu
-* [MNT] Add the code coverage on codecov | @xuyxu
-* [MNT] Add the version numbers to requirements.txt | @zackhardtoname and @xuyxu
+* |MajorFeature| Add :meth:`set_scheduler` for all ensembles | @xuyxu
+* |MajorFeature| Add :class:`AdversarialTrainingClassifier` and :class:`AdversarialTrainingRegressor` | @xuyxu
+* |MajorFeature| Add :class:`SnapshotEnsembleClassifier` and :class:`SnapshotEnsembleRegressor` | @xuyxu
+* |MajorFeature| Add model validation and serialization | @ghost-ronin and @xuyxu
+* |Enhancement| Add CI and maintenance tools | @xuyxu
+* |Enhancement| Add the code coverage on codecov | @xuyxu
+* |Enhancement| Add the version numbers to requirements.txt | @zackhardtoname and @xuyxu
+* |Enhancement| Refactor the codes on operating tensors into an independent module | @zzzzwj
+* |Enhancement| Refactor the set_optimizer into an independent method | @xuyxu
+* |Enhancement| Improve the logging module | @zzzzwj
+* |API| Remove the input argument ``output_dim`` from all methods | @xuyxu
+* |Fix| Fix the bug in logging module when using multi-processing | @zzzzwj
 
-Changed
-~~~~~~~
-* [ENH] Refactor the codes on operating tensors into an independent module | @zzzzwj
-* [ENH] Refactor the set_optimizer into an independent method | @xuyxu
-* [ENH] Improve the logging module | @zzzzwj
 
-Removed
-~~~~~~~
-* [ENH] Remove the input parameter `output_dim` from all methods | @xuyxu
+.. role:: raw-html(raw)
+   :format: html
 
-Fixed
-~~~~~
-* [BUG] Fix the bug in logging module when using multi-processing | @zzzzwj
+.. role:: raw-latex(raw)
+   :format: latex
+
+.. |MajorFeature| replace:: :raw-html:`<span class="badge badge-success">Major Feature</span>` :raw-latex:`{\small\sc [Major Feature]}`
+.. |Feature| replace:: :raw-html:`<span class="badge badge-success">Feature</span>` :raw-latex:`{\small\sc [Feature]}`
+.. |Efficiency| replace:: :raw-html:`<span class="badge badge-info">Efficiency</span>` :raw-latex:`{\small\sc [Efficiency]}`
+.. |Enhancement| replace:: :raw-html:`<span class="badge badge-info">Enhancement</span>` :raw-latex:`{\small\sc [Enhancement]}`
+.. |Fix| replace:: :raw-html:`<span class="badge badge-danger">Fix</span>` :raw-latex:`{\small\sc [Fix]}`
+.. |API| replace:: :raw-html:`<span class="badge badge-warning">API Change</span>` :raw-latex:`{\small\sc [API Change]}`

docs/parameters.rst

@@ -120,6 +120,9 @@ Reference:
     B. Lakshminarayanan, A. Pritzel, C. Blundell., Simple and Scalable
     Predictive Uncertainty Estimation using Deep Ensembles, NIPS 2017.
 
+.. warning::
+    When your base estimator is under-fit on the dataset, it is not recommended to use the :mod:`AdversarialTrainingClassifier` or :mod:`AdversarialTrainingRegressor`, because they may deteriorate the performance further.
+
 AdversarialTrainingClassifier
 *****************************
 

docs/quick_start.rst

@@ -15,6 +15,9 @@ You can use the latest version of Ensemble-PyTorch with the following command:
 
 Ensemble-PyTorch is designed to be portable and has very small package dependencies. It is recommended to use the Python environment and PyTorch installed from `Anaconda <https://www.anaconda.com/>`__. In this case, there is no need to run the third command in the code snippet above.
 
+.. tip::
+    Ensemble-PyTorch is close to the end of the Beta stage. It will soon be uploaded to `PyPI <https://pypi.org/>`__.
+
 Define Your Base Estimator
 --------------------------
 
@@ -62,8 +65,8 @@ Ensemble-PyTorch uses a global logger to track and print the intermediate inform
 
 Using the logger, all intermediate information will be printed on the command line and saved to the specified text file: classification_mnist_mlp.
 
-Choose the Ensemble Method
---------------------------
+Choose the Ensemble
+-------------------
 
 After defining the base estimator, we can then wrap it using one of ensemble methods available in Ensemble-PyTorch. Different methods have very similar APIs, take the ``VotingClassifier`` as an example:
 

docs/requirements.txt

@@ -1,3 +1,4 @@
 sphinx==3.2.*
+sphinx-panels==0.5.*
 sphinx-copybutton
 guzzle_sphinx_theme

docs/roadmap.rst

@@ -1,2 +1,8 @@
 Roadmap
 =======
+
+[0.1.0] - ~
+-----------
+
+* Add the partial-fit mode for all ensembles.
+* Support manually-specified criteria in the ``predict`` function.