Support PyMC v5 (#309)

* Implementing support for pymc v4

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* workflow error

* workflow

* allow python version for lint

* getting distributions to work

* some tweaks for pymc3

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* pymc3 fixes

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* refactoring distributions again

* docstrings for distributions

* starting to re-write docs

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* lint and starting to update tutorials

* fixing some eccentricity dist issues

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* updating docs

* updating citations

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* update README [ci skip]

* making orphan page [ci skip]

* Require PyMC < 5 for PyMC v4 version of exoplanet

* Require pymc >= 5 in setup.py

Left TODO comments about other exoplanet-dev packages. No released version yet but their main branch has support for pymc >= 5

* Replace aesara by pytensor and pymc4 by pymc

The actual imports are in the compat module, but I updated `at` to `pt`.
That way `exoplanet` stays up to date with latest "notation" for `pytensor.tensor`

* Add upper limit on xarray for pymc3

Causes numpy error for Python >= 3.9 otherwise

* Accept pre-commit changes on tutorial notebooks

* Replace `AESARA_FLAGS` by `PYTENSOR_FLAGS` in `pytest.ini`

* Update docs from PyMC 4 to PyMC 5 (mostly pytensor -> Aesara)

* Comment-out warnings in `docs_setup()` with TODO to maybe remove them

* TODO comment for deprecated `pt.opt.Assert`

* Update `Evaluator()` call to use `model` kwarg

* Replace testval by initval and start by initvals

* Enable using kipping13 prior when eccentricity is a derived parameter

In PyMC 3, we could just used `observed=ecc` with any distribution.
In PyMC >= 5, we need to use a `pm.Potential`

* Add `_truncate_dist` helper function for eccentricity prior

* Fix vanelyen19 prior when eccentricity is a derived parameter

Same as kipping13: if eccentricity is "observed" (derived), extra prior must be a `pm.Potential`

* Test `angle()` and `unit_disk()` distributions and fix shape error

Previously, default initval was hardcoded so passing shape without initval caused error. Added failing tests and fixed

* Pop initval when ecc is None only when bounds are set

* Clean-up `_truncate_dist()` helper function

* Raise `NotImplementedError` when trying to pass observed ecc with bounds in kipping13

This did not work with the original PyMC3 implementation either, but is supported in PyMC 5.
This error makes it a bit clearer

* Test eccentricity priors with and without bounds

* Update required Python version from 3.6 to 3.8 in docs

* Convert shape from `int` to `tuple` for pymc>3 in impact parameter distribution

* Remove deprecated `pt.opt.Assert` in favor of `Assert` from `compat`

* Remove leftover commented-out warning filters

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Replace tox dependency by nox

* Check that kipping13_observed raises error with bounds when using PyMC3

* Get shape using `tag.test_value` for PyMC3 tests

`type.shape` works only in PyMC > 3

* Remove extra potential in vaneylen19 distribution PyMC 3 version

The implementation that worked with PyMC v5 did not in PyMC3

* Update dependency versions for PyMC v5

* Add `exoplanet-core` dependency to `docs`

Had to remove from the main dependencies because no of the different versions for PyMC3 and PyMC 5

* Another merge conflict

* Range of tests for deps

* Don't install coverage

---------

Co-authored-by: Dan F-M <foreman.mackey@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Author: 3 people

.clang-format

@@ -1,6 +1,5 @@
 paper/* linguist-documentation
 docs/* linguist-documentation
-exoplanet/theano_ops/vendor/* linguist-vendored
 *.h linguist-language=C++
 docs/notebooks/*.ipynb filter=nbstripout
 paper/figures/*.ipynb filter=nbstripout

.gitattributes

@@ -13,76 +13,46 @@ on:
 
 jobs:
   tests:
-    runs-on: ubuntu-latest
+    runs-on: ${{ matrix.os }}
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["3.9", "3.10"]
+        python-version: ["3.10"]
+        os: ["ubuntu-latest"]
+        session: ["test_pymc3", "test_pymc"]
+        include:
+          - python-version: "3.9"
+            os: "ubuntu-latest"
+            session: "test_pymc3"
+          - python-version: "3.11"
+            os: "ubuntu-latest"
+            session: "test_pymc"
+          - python-version: "3.10"
+            os: "ubuntu-latest"
+            session: "lint"
 
     steps:
       - name: Checkout
         uses: actions/checkout@v4
         with:
           fetch-depth: 0
 
-      - name: Setup Python
-        uses: actions/setup-python@v5
+      - uses: actions/setup-python@v4
+        name: Install Python
         with:
           python-version: ${{ matrix.python-version }}
 
       - name: Install dependencies
         run: |
           python -m pip install -U pip
-          python -m pip install -U coveralls coverage[toml] tox tox-gh-actions
+          python -m pip install -U nox
 
       - name: Run tests
-        run: python -m tox
-
-      - name: Combine and upload coverage
-        run: |
-          python -m coverage combine
-          python -m coverage xml -i
-          python -m coveralls --service=github
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-          COVERALLS_PARALLEL: true
-          COVERALLS_FLAG_NAME: py${{ matrix.python-version }}
-
-  coverage:
-    needs: tests
-    runs-on: ubuntu-latest
-    steps:
-      - name: Setup Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.9"
-      - name: Finish coverage collection
-        run: |
-          python -m pip install -U pip
-          python -m pip install -U coveralls
-          python -m coveralls --finish
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-
-  lint:
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v4
-        with:
-          fetch-depth: 0
-
-      - name: Setup Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.9"
-
-      - name: Install dependencies
         run: |
-          python -m pip install -U pip
-          python -m pip install tox
-
-      - name: Lint the code
-        run: python -m tox -e lint
+          python -m nox --non-interactive \
+            --error-on-missing-interpreter \
+            --python ${{ matrix.python-version }} \
+            --session ${{ matrix.session }}
 
   build:
     runs-on: ubuntu-latest
@@ -104,7 +74,7 @@ jobs:
           path: dist/*
 
   upload_pypi:
-    needs: [tests, lint, build]
+    needs: [tests, build]
     runs-on: ubuntu-latest
     if: startsWith(github.ref, 'refs/tags/')
     steps:

.github/workflows/tests.yml

@@ -17,10 +17,11 @@
 Fast & scalable MCMC for all your exoplanet needs! _exoplanet_ is a toolkit for
 probabilistic modeling of time series data in astronomy with a focus on
 observations of [exoplanets](https://en.wikipedia.org/wiki/Exoplanet), using
-[PyMC3](https://docs.pymc.io). _PyMC3_ is a flexible and high-performance model
-building language and inference engine that scales well to problems with a large
-number of parameters. _exoplanet_ extends _PyMC3_'s language to support many of
-the custom functions and distributions required when fitting exoplanet datasets.
+[PyMC](https://www.pymc.io). _PyMC_ is a flexible and high-performance
+model-building language and inference engine that scales well to problems with a
+large number of parameters. _exoplanet_ extends _PyMC_'s language to support
+many of the custom functions and distributions required when fitting exoplanet
+datasets.
 
 Read the full documentation at [docs.exoplanet.codes](https://docs.exoplanet.codes).
 
@@ -29,7 +30,7 @@ Read the full documentation at [docs.exoplanet.codes](https://docs.exoplanet.cod
 The quickest way to get started is to use [pip](https://pip.pypa.io):
 
 ```bash
-python -m pip install exoplanet
+python -m pip install "exoplanet[pymc]"
 ```
 
 Note that you will need Python (>=3.6) installed for this to work, but then this
@@ -47,7 +48,7 @@ page](https://gallery.exoplanet.codes).
 
 ## Contributing
 
-_exoplanet_ is an open source project and we would love it if you wanted to
+_exoplanet_ is an open-source project, and we would love it if you wanted to
 contribute. Check out [the developer
 documentation](https://docs.exoplanet.codes/en/latest/user/dev/) for more info
 about getting started.

README.md

@@ -8,4 +8,4 @@ pymc3-ext>=0.1.0
 emcee
 numpy<1.22
 xarray<2023.10.0
-.
+.[docs]

binder/requirements.txt

@@ -3,10 +3,10 @@ exoplanet
 
 *exoplanet* is a toolkit for probabilistic modeling of time series data in
 astronomy with a focus on observations of `exoplanets
-<https://en.wikipedia.org/wiki/Exoplanet>`_, using `PyMC3
-<https://docs.pymc.io>`_. *PyMC3* is a flexible and high-performance model
+<https://en.wikipedia.org/wiki/Exoplanet>`_, using `PyMC
+<https://www.pymc.io>`_. *PyMC* is a flexible and high-performance model
 building language and inference engine that scales well to problems with a large
-number of parameters. *exoplanet* extends *PyMC3*'s language to support many of
+number of parameters. *exoplanet* extends *PyMC*'s language to support many of
 the custom functions and distributions required when fitting exoplanet datasets.
 These features include:
 
@@ -44,13 +44,13 @@ an issue <https://github.com/exoplanet-dev/exoplanet/issues>`_ there.
    🖼 For more in depth examples of *exoplanet* used for more realistic problems,
    go to the `Case studies page <https://gallery.exoplanet.codes>`_.
 
-   📈 For more information about scalable Gaussian Processes in PyMC3 (this was
+   📈 For more information about scalable Gaussian Processes in PyMC (this was
    previously implemented as part of *exoplanet*), see the `celerite2 documentation
    page <https://celerite2.readthedocs.io>`_.
 
-   👉 For helper functions and PyMC3 extras that used to be implemented as part of
-   *exoplanet*, see the `pymc3-ext project
-   <https://github.com/exoplanet-dev/pymc3-ext>`_.
+   👉 For helper functions and PyMC extras that used to be implemented as part of
+   *exoplanet*, see the `pymc-ext project
+   <https://github.com/exoplanet-dev/pymc-ext>`_.
 
 
 Contents
@@ -62,7 +62,6 @@ Contents
 
    user/install
    tutorials/citation.md
-   user/theano
    user/multiprocessing
    user/api
    user/dev
@@ -74,7 +73,7 @@ Contents
 
    tutorials/about.md
    tutorials/autodiff.md
-   tutorials/intro-to-pymc3.md
+   tutorials/intro-to-pymc.md
    tutorials/data-and-models.md
    tutorials/light-delay.md
    tutorials/reparameterization.md

docs/index.rst

@@ -6,7 +6,7 @@ jupytext:
     format_version: 0.13
     jupytext_version: 1.15.2
 kernelspec:
-  display_name: Python 3
+  display_name: Python 3 (ipykernel)
   language: python
   name: python3
 ---
@@ -15,8 +15,8 @@ kernelspec:
 
 +++
 
-This and the following tutorials are automatically executed with every change of the code to make sure that they are always up to date with the code.
-As a result, they are designed to require only a relatively small amount of computation time; when using `exoplanet` for research you will probably find that your runtimes are longer.
+This and the following tutorials are automatically executed with every change of the code to make sure that they are always up-to-date with the code.
+As a result, they are designed to require only a relatively small amount of computation time; when using `exoplanet` for research you will probably find that your run times are longer.
 For more in-depth tutorials with real-world applications and real data, check out the [Case Studies page](https://gallery.exoplanet.codes).
 
 At the top of each tutorial, you'll find a cell like the following that indicates the version of `exoplanet` that was used to generate the tutorial:
@@ -28,9 +28,9 @@ exoplanet.utils.docs_setup()
 print(f"exoplanet.__version__ = '{exoplanet.__version__}'")
 ```
 
-That cell also includes a call to the `exoplanet.utils.docs_setup` function that will squash some warnings (these are generally caused by Theano/Aesara; see {ref}`theano` for more info) and set up our `matplotlib` style.
+That cell also includes a call to the `exoplanet.utils.docs_setup` function that will squash some warnings (these are generally caused by Theano/PyTensor) and set up our `matplotlib` style.
 
-To exectute a tutorial on your own, you can click on the buttons at the top right or this page to launch the notebook using [Binder](https://mybinder.org) or download the `.ipynb` file directly.
+To execute a tutorial on your own, you can click on the buttons at the top right or this page to launch the notebook using [Binder](https://mybinder.org) or download the `.ipynb` file directly.
 
 ```{code-cell}