Merge branch 'cdf_methods' into cdf_methods

Author: fonnesbeck

.travis.yml

@@ -6,19 +6,18 @@ before_install:
   - "sh -e /etc/init.d/xvfb start"
   - if [ ${MATPLOTLIB} = "1.2" ]; then mkdir $HOME/.matplotlib; fi
   - if [ ${MATPLOTLIB} = "1.2" ]; then cp ${SRCDIR}/tools/matplotlibrc $HOME/.matplotlib/matplotlibrc; fi
-  #- mkdir -p $HOME/.config/matplotlib; echo "backend: agg" > $HOME/.config/matplotlib/matplotlibrc
 
 install:
   - . ./scripts/create_testenv.sh
   - pip install coveralls pylint
 
 env:
-  - PYTHON_VERSION=2.7 TESTCMD=" -vv --with-timer --with-coverage --cover-package=pymc3 -e test_examples -e test_distributions"
-  - PYTHON_VERSION=2.7 RUN_PYLINT="true" TESTCMD=" -vv --with-timer --with-coverage --cover-package=pymc3 pymc3.tests.test_distributions pymc3.tests.test_examples"
-  - PYTHON_VERSION=2.7 TESTCMD=" -vv --with-timer --with-coverage --cover-package=pymc3 pymc3.tests.test_distributions_random"
-  - PYTHON_VERSION=3.6 TESTCMD=" -vv --with-timer --with-coverage --cover-package=pymc3 -e test_examples -e test_distributions"
-  - PYTHON_VERSION=3.6 TESTCMD=" -vv --with-timer --with-coverage --cover-package=pymc3 pymc3.tests.test_distributions pymc3.tests.test_examples"
-  - PYTHON_VERSION=3.6 TESTCMD=" -vv --with-timer --with-coverage --cover-package=pymc3 pymc3.tests.test_distributions_random"
+  - PYTHON_VERSION=2.7 TESTCMD="--durations=10 --ignore=pymc3/tests/test_examples.py --cov-append --ignore=pymc3/tests/test_distributions_random.py --ignore=pymc3/tests/test_variational_inference.py --ignore=pymc3/tests/test_shared.py --ignore=pymc3/tests/test_smc.py"
+  - PYTHON_VERSION=2.7 RUN_PYLINT="true" TESTCMD="--durations=10 --cov-append pymc3/tests/test_distributions_random.py pymc3/tests/test_shared.py pymc3/tests/test_smc.py"
+  - PYTHON_VERSION=2.7 TESTCMD="--durations=10 --cov-append pymc3/tests/test_examples.py pymc3/tests/test_variational_inference.py"
+  - PYTHON_VERSION=3.6 TESTCMD="--durations=10 --cov-append --ignore=pymc3/tests/test_examples.py --ignore=pymc3/tests/test_distributions_random.py --ignore=pymc3/tests/test_variational_inference.py --ignore=pymc3/tests/test_shared.py --ignore=pymc3/tests/test_smc.py"
+  - PYTHON_VERSION=3.6 TESTCMD="--durations=10 --cov-append pymc3/tests/test_distributions_random.py pymc3/tests/test_shared.py pymc3/tests/test_smc.py"
+  - PYTHON_VERSION=3.6 TESTCMD="--durations=10 --cov-append pymc3/tests/test_examples.py pymc3/tests/test_variational_inference.py"
 script:
   - . ./scripts/test.sh $TESTCMD
 

CONTRIBUTING.md

@@ -74,14 +74,16 @@ We recommended that your contribution complies with the following guidelines bef
 
 * Documentation and high-coverage tests are necessary for enhancements to be accepted.
 
+* Run any of the pre-existing examples in ``docs/source/notebooks`` that contain analyses that would be affected by your changes to ensure that nothing breaks. This is a useful opportunity to not only check your work for bugs that might not be revealed by unit test, but also to show how your contribution improves PyMC3 for end users.
+
 You can also check for common programming errors with the following
 tools:
 
-* Code with good unittest **coverage** (at least 80%), check with:
+* Code with good test **coverage** (at least 80%), check with:
 
   ```bash
-  $ pip install nose coverage
-  $ nosetests --with-coverage path/to/tests_for_package
+  $ pip install pytest pytest-cov coverage
+  $ pytest --cov=pymc3 pymc3/tests/tests_for_package.py
   ```
 
 * No `pyflakes` warnings, check with:
@@ -110,9 +112,12 @@ tools:
 We have provided a Dockerfile which helps for isolating build problems, and local development.
 Install [Docker](https://www.docker.com/) for your operating system, clone this repo, then
 run `./scripts/start_container.sh`. This should start a local docker container called `pymc3`,
-as well as a [`jupyter`](http://jupyter.org/) notebook server running on port 8888. You will have to open
-a browser at `localhost:8888`. The repo will be running the code from your local copy of `pymc3`, 
-so it is good for development.  You may also use it to run the test suite, with
+as well as a [`jupyter`](http://jupyter.org/) notebook server running on port 8888. The
+notebook should be opened in your browser automatically (you can disable this by passing
+`--no-browser`). The repo will be running the code from your local copy of `pymc3`,
+so it is good for development.  To find the token necessary to access the notebooks, run
+
+You may also use it to run the test suite, with
 
 ```bash
 $  docker exec -it pymc3  bash # logon to the container
@@ -122,6 +127,13 @@ $  . ./scripts/test.sh # takes a while!
 
 This should be quite close to how the tests run on TravisCI.
 
+If the container was started without opening the browser, you
+need the a token to work with the notebook. This token can be
+access with
+
+```
+docker exec -it pymc3 jupyter notebook list
+```
 
 ## Style guide
 

MANIFEST.in

@@ -1,5 +1,6 @@
 recursive-include pymc3/examples/data *
 recursive-include source *
+recursive-include docs *
 include requirements.txt
 include *.md *.rst
-
+include scripts/*.sh

README.rst

@@ -46,6 +46,7 @@ Getting started
 -  `PyMC3 port of the models presented in the book "Doing Bayesian Data
    Analysis" by John
    Kruschke <https://github.com/aloctavodia/Doing_bayesian_data_analysis>`__
+-  `PyMC3 port of the models presented in the book "Bayesian Cognitive Modeling" by Michael Lee and EJ Wagenmakers <https://github.com/junpenglao/Bayesian-Cognitive-Modeling-in-Pymc3>`__
 -  `Coyle P. (2016) Probabilistic programming and PyMC3. European Scientific Python Conference 2015 (Cambridge, UK) <http://adsabs.harvard.edu/abs/2016arXiv160700379C>`__
 -  `Bayesian Analysis with Python by Osvaldo Martin <https://www.packtpub.com/big-data-and-business-intelligence/bayesian-analysis-python>`__ (and `errata <https://github.com/aloctavodia/BAP>`__)
 

RELEASE-NOTES.md

@@ -4,9 +4,31 @@
 
 ### New features
 
-* Experimental GPU support.
+* [Add Gaussian Process submodule](http://pymc-devs.github.io/pymc3/notebooks/GP-introduction.html)
 
-* Gaussian Process support.
+* Many optimizations and speed-ups.
+
+* NUTS implementation now matches current Stan implementation.
+
+* Add higher-order integrators for HMC.
+
+* [Add sampler statistics.](http://pymc-devs.github.io/pymc3/notebooks/sampler-stats.html)
+
+* [Add live-trace to see samples in real-time.](http://pymc-devs.github.io/pymc3/notebooks/live_sample_plots.html)
+
+* ADVI stopping criterion implemented.
+
+* Improved support for theano's floatX setting to enable GPU computatiosn (experimental).
+
+* MvNormal supports Cholesky Decomposition now for increased speed and numerical stability.
+
+* [Add Elliptical Slice Sampler.](http://pymc-devs.github.io/pymc3/notebooks/GP-slice-sampling.html)
+
+* Add Stein-Variational Gradient Descent (experimental).
+
+* [Add Operator Variational Inference (experimental).](http://pymc-devs.github.io/pymc3/notebooks/bayesian_neural_network_opvi-advi.html).
+
+* [Add pm.generator() to easily specify mini-batches.](http://pymc-devs.github.io/pymc3/notebooks/bayesian_neural_network_opvi-advi.html#Minibatch-ADVI)
 
 * `Model` can now be inherited from and act as a base class for user specified models (see pymc3.models.linear).
 
@@ -16,6 +38,8 @@
 
 * GLM models do not need a left-hand variable anymore.
 
+* Add support for cholesky parametrizations for cov/corr matrices.
+
 * Refactored HMC and NUTS for better readability.
 
 * Add support for Python 3.6.