- Gaussian process Regressor

- Student's t process Regressor

- Sparse Gaussian process Regressor

Thank you for considering contributing to ``pymc-learn``! This project is intended to be a space where anyone can share models they've built.

Please read these guidelines before submitting anything to the project. As of the first release, I'm the only person working on this project so respecting these guidelines will help me get back to you more quickly.

- Something I haven't thought of?

To create a Pull Request against this library, please fork the project and work from there.

Always use a ``feature`` branch. It's good practice to never routinely work on the ``master`` branch.

Transitioning from PyMC3 to PyMC4

.. raw:: html

<embed>

<blockquote class="twitter-tweet" data-lang="en"><p lang="en" dir="ltr">.<a href="https://twitter.com/pymc_learn?ref_src=twsrc%5Etfw">@pymc_learn</a> has been following closely the development of <a href="https://twitter.com/hashtag/PyMC4?src=hash&amp;ref_src=twsrc%5Etfw">#PyMC4</a> with the aim of switching its backend from <a href="https://twitter.com/hashtag/PyMC3?src=hash&amp;ref_src=twsrc%5Etfw">#PyMC3</a> to PyMC4 as the latter grows to maturity. Core devs are invited. Here&#39;s the tentative roadmap for PyMC4: <a href="https://t.co/Kwjkykqzup">https://t.co/Kwjkykqzup</a> cc <a href="https://twitter.com/pymc_devs?ref_src=twsrc%5Etfw">@pymc_devs</a> <a href="https://t.co/Ze0tyPsIGH">https://t.co/Ze0tyPsIGH</a></p>&mdash; pymc-learn (@pymc_learn) <a href="https://twitter.com/pymc_learn/status/1059474316801249280?ref_src=twsrc%5Etfw">November 5, 2018</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

</embed>

and `PyMC3 <https://docs.pymc.io/>`_ and is distributed under the new BSD-3 license,

provided by `PyMC3 <https://docs.pymc.io/>`_.

.. NOTE::

``pymc-learn`` leverages and extends the Base template provided by the

PyMC3 Models project: https://github.com/parsing-science/pymc3_models

Transitioning from PyMC3 to PyMC4

.. raw:: html

<embed>

<blockquote class="twitter-tweet" data-lang="en"><p lang="en" dir="ltr">.<a href="https://twitter.com/pymc_learn?ref_src=twsrc%5Etfw">@pymc_learn</a> has been following closely the development of <a href="https://twitter.com/hashtag/PyMC4?src=hash&amp;ref_src=twsrc%5Etfw">#PyMC4</a> with the aim of switching its backend from <a href="https://twitter.com/hashtag/PyMC3?src=hash&amp;ref_src=twsrc%5Etfw">#PyMC3</a> to PyMC4 as the latter grows to maturity. Core devs are invited. Here&#39;s the tentative roadmap for PyMC4: <a href="https://t.co/Kwjkykqzup">https://t.co/Kwjkykqzup</a> cc <a href="https://twitter.com/pymc_devs?ref_src=twsrc%5Etfw">@pymc_devs</a> <a href="https://t.co/Ze0tyPsIGH">https://t.co/Ze0tyPsIGH</a></p>&mdash; pymc-learn (@pymc_learn) <a href="https://twitter.com/pymc_learn/status/1059474316801249280?ref_src=twsrc%5Etfw">November 5, 2018</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

</embed>

``pymc-learn`` is tested on Python 2.7, 3.5 & 3.6 and depends on Theano,

PyMC3, Scikit-learn, NumPy, SciPy, and Matplotlib (see ``requirements.txt``

an optimization problem. ADVI is provided PyMC3.

If you want to cite ``pymc-learn`` for its API, you may also want to consider

this reference::

Or using BibTex as follows:

.. code-block:: latex

@article{Pymc3_models,

title={pymc3_models: Custom PyMC3 models built on top of the scikit-learn API,

author={Carlson, Nicole},

journal={},

url={https://github.com/parsing-science/pymc3_models}

year={2018}

}

``pymc-learn`` leverages and extends the Base template provided by the PyMC3

Models project: https://github.com/parsing-science/pymc3_models.

0.0.1 / 2019-01-01

- Added Gaussian process regression, students t process, sparse gaussian process

- Added Multilayer perceptron

Thank you for considering contributing to ``pymc-learn``! This project is intended to be a space where anyone can share models they've built.

Please read these guidelines before submitting anything to the project. As of the first release, I'm the only person working on this project so respecting these guidelines will help me get back to you more quickly.

- Something I haven't thought of?

- Lines can be 120 characters long

Finally, if you need to get in touch for information about the project, `send us an e-mail <devs@pymc-learn.org>`__.

Transitioning from PyMC3 to PyMC4

.. raw:: html

<embed>

<blockquote class="twitter-tweet" data-lang="en"><p lang="en" dir="ltr">.<a href="https://twitter.com/pymc_learn?ref_src=twsrc%5Etfw">@pymc_learn</a> has been following closely the development of <a href="https://twitter.com/hashtag/PyMC4?src=hash&amp;ref_src=twsrc%5Etfw">#PyMC4</a> with the aim of switching its backend from <a href="https://twitter.com/hashtag/PyMC3?src=hash&amp;ref_src=twsrc%5Etfw">#PyMC3</a> to PyMC4 as the latter grows to maturity. Core devs are invited. Here&#39;s the tentative roadmap for PyMC4: <a href="https://t.co/Kwjkykqzup">https://t.co/Kwjkykqzup</a> cc <a href="https://twitter.com/pymc_devs?ref_src=twsrc%5Etfw">@pymc_devs</a> <a href="https://t.co/Ze0tyPsIGH">https://t.co/Ze0tyPsIGH</a></p>&mdash; pymc-learn (@pymc_learn) <a href="https://twitter.com/pymc_learn/status/1059474316801249280?ref_src=twsrc%5Etfw">November 5, 2018</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

</embed>

This also installs required dependencies including Theano.

For alternative Theano installations (e.g., gpu), please see the

instructions on the main `Theano webpage <http://deeplearning.net/software/theano/>`_.

Transitioning from PyMC3 to PyMC4

.. raw:: html

<embed>

<blockquote class="twitter-tweet" data-lang="en"><p lang="en" dir="ltr">.<a href="https://twitter.com/pymc_learn?ref_src=twsrc%5Etfw">@pymc_learn</a> has been following closely the development of <a href="https://twitter.com/hashtag/PyMC4?src=hash&amp;ref_src=twsrc%5Etfw">#PyMC4</a> with the aim of switching its backend from <a href="https://twitter.com/hashtag/PyMC3?src=hash&amp;ref_src=twsrc%5Etfw">#PyMC3</a> to PyMC4 as the latter grows to maturity. Core devs are invited. Here&#39;s the tentative roadmap for PyMC4: <a href="https://t.co/Kwjkykqzup">https://t.co/Kwjkykqzup</a> cc <a href="https://twitter.com/pymc_devs?ref_src=twsrc%5Etfw">@pymc_devs</a> <a href="https://t.co/Ze0tyPsIGH">https://t.co/Ze0tyPsIGH</a></p>&mdash; pymc-learn (@pymc_learn) <a href="https://twitter.com/pymc_learn/status/1059474316801249280?ref_src=twsrc%5Etfw">November 5, 2018</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

</embed>

- *Flexibility*: This ease of use does not come at the cost of reduced flexibility. Given that ``pymc-learn`` integrates with `PyMC3 <https://docs.pymc.io>`_, it enables you to implement anything you could have built in the base language.

- *Performance*. The primary inference algorithm is gradient-based automatic differentiation variational inference (ADVI) (Kucukelbir et al., 2017), which estimates a divergence measure between approximate and true posterior distributions. Pymc-learn scales to complex, high-dimensional models thanks to GPU-accelerated tensor math and reverse-mode automatic differentiation via Theano (Theano Development Team, 2016), and it scales to large datasets thanks to estimates computed over mini-batches of data in ADVI.

The underlying probabilistic models are built using pymc3 (Salvatier et al., 2016).

Transitioning from PyMC3 to PyMC4

.. raw:: html

<embed>

<blockquote class="twitter-tweet" data-lang="en"><p lang="en" dir="ltr">.<a href="https://twitter.com/pymc_learn?ref_src=twsrc%5Etfw">@pymc_learn</a> has been following closely the development of <a href="https://twitter.com/hashtag/PyMC4?src=hash&amp;ref_src=twsrc%5Etfw">#PyMC4</a> with the aim of switching its backend from <a href="https://twitter.com/hashtag/PyMC3?src=hash&amp;ref_src=twsrc%5Etfw">#PyMC3</a> to PyMC4 as the latter grows to maturity. Core devs are invited. Here&#39;s the tentative roadmap for PyMC4: <a href="https://t.co/Kwjkykqzup">https://t.co/Kwjkykqzup</a> cc <a href="https://twitter.com/pymc_devs?ref_src=twsrc%5Etfw">@pymc_devs</a> <a href="https://t.co/Ze0tyPsIGH">https://t.co/Ze0tyPsIGH</a></p>&mdash; pymc-learn (@pymc_learn) <a href="https://twitter.com/pymc_learn/status/1059474316801249280?ref_src=twsrc%5Etfw">November 5, 2018</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

</embed>

Why scikit-learn and PyMC3

PyMC3 is a Python package for probabilistic machine learning that enables users

modeling framework. However, PyMC3 lacks the steps between creating a model and

reusing it with new data in production. The missing steps include: scoring a

model, saving a model for later use, and loading the model in production

systems.

In contrast, *scikit-learn* which has become the standard

users to train, score, save and load models in production. However,

*scikit-learn* may not have the model for a user's specific problem.

These limitations have led to the development of the open

source *pymc3-models* library which provides a template to build bespoke

PyMC3 models on top of the *scikit-learn* API and reuse them in

production. This enables users to easily and quickly train, score, save and

load their bespoke models just like in *scikit-learn*.

The ``pymc-learn`` project adopted and extended the template in *pymc3-models*

to develop probabilistic versions of the estimators in *scikit-learn*.

5. Salvatier, J., Wiecki, T. V., & Fonnesbeck, C. (2016). Probabilistic programming in Python using PyMC3. PeerJ Computer Science, 2, e55.

8. Theano Development Team. Theano: A Python framework for fast computation of mathematical expressions. arXiv e-prints, abs/1605.02688, May 2016. URL http://arxiv.org/abs/1605.02688.