qpytorch 0.1__tar.gz → 0.1.1__tar.gz

{qpytorch-0.1 → qpytorch-0.1.1}/.conda/meta.yaml

@@ -24,9 +24,10 @@ requirements:
     - jaxtyping
     - linear_operator>=0.6
     - mpmath>=0.19,<=1.3
-    - pytorch>=2.2
+    - pytorch>=2.0
     - scikit-learn
     - gpytorch>=1.13
+    - numpy<2
 
 test:
   imports:

{qpytorch-0.1 → qpytorch-0.1.1}/CONTRIBUTING.md

@@ -6,9 +6,10 @@ Thanks for contributing!
 
 To get the development installation with all the necessary dependencies for
 linting, testing, and building the documentation, run the following:
+
 ```bash
 git clone https://github.com/lanzithinking/qepytorch.git
-cd qpytorch
+cd qepytorch
 pip install -e .[dev,docs,examples,keops,pyro,test]  # keops and pyro are optional
 pre-commit install
 ```
@@ -25,6 +26,7 @@ The [development installation instructions](#development-installation) should in
 
 `flake8` and `pre-commit` are both run every time you make a local commit.
 To run both commands independent of making a commit:
+
 ```bash
 SKIP=flake8 pre-commit run --files test/**/*.py qpytorch/**/*.py
 flake8
@@ -36,7 +38,7 @@ We use [standard sphinx docstrings](https://sphinx-rtd-tutorial.readthedocs.io/e
 
 ### Type Hints
 
-Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch aims to be fully typed using Python 3.8+
+Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch aims to be fully typed using Python 3.10+
 [type hints](https://www.python.org/dev/peps/pep-0484/).
 
 We recognize that we have a long way to go towards fully typing the library,
@@ -54,7 +56,7 @@ python -m unittest
 ```
 
 - To run tests within a specific directory, run (e.g.) `python -m unittest discover -s test/kernels`.
-- To run a specific unit test, run (e.g.) `python -m unittest test.operators.test_rbf_kernel.TestRBFKernel.test_active_dims_range`.
+- To run a specific unit test, run (e.g.) `python -m unittest test.distributions.test_qexponential.TestQExponential.test_batch_log_prob`.
 
 
 ### Documentation
@@ -121,8 +123,10 @@ We accept the following types of issues:
 - Opportuntities to refactor code
 - Performance issues (speed, memory, etc.)
 
+<!--
 Please refrain from using the issue tracker for questions or debugging personal code.
-Instead please use the [Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch discussions forum](https://github.com/lanzithinking/qepytorch/discussions).
+Instead please use the [QePyTorch discussions forum](https://github.com/lanzithinking/qepytorch/discussions).
+-->
 
 ## License
 

{qpytorch-0.1 → qpytorch-0.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: qpytorch
-Version: 0.1
+Version: 0.1.1
 Summary: An implementation of Q-Exponential Processes in Pytorch based on GPyTorch
 Home-page: https://lanzithinking.github.io/qepytorch/
 Author: Shiwei Lan
@@ -28,60 +28,61 @@ License-File: LICENSE
 [![License](https://img.shields.io/badge/license-MIT-green.svg)](LICENSE)
 
 [![Python Version](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/)
-[![Conda](https://img.shields.io/conda/v/conda-forge/qpytorch.svg)](https://anaconda.org/conda-forge/qpytorch)
+[![Conda](https://img.shields.io/conda/v/qpytorch.svg)](https://anaconda.org/qpytorch)
 [![PyPI](https://img.shields.io/pypi/v/qpytorch.svg)](https://pypi.org/project/qpytorch)
 
-Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is a Q-exponential process library implemented using PyTorch built on [GPyTorch](https://gpytorch.ai). Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is designed for creating scalable, flexible, and modular Q-exponential process models with ease.
+Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is a Python package for Q-exponential process ([QEP](https://papers.nips.cc/paper_files/paper/2023/file/e6bfdd58f1326ff821a1b92743963bdf-Paper-Conference.pdf)) implemented using PyTorch and built up on [GPyTorch](https://gpytorch.ai). Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is designed to facilitate creating scalable, flexible, and modular QPE models.
 
-Internally, Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch differs from many existing approaches to QEP inference by performing most inference operations using numerical linear algebra techniques like preconditioned conjugate gradients.
-Implementing a scalable QEP method is as simple as providing a matrix multiplication routine with the kernel matrix and its derivative via our [LinearOperator](https://github.com/cornellius-gp/linear_operator) interface,
-or by composing many of our already existing `LinearOperators`.
-This allows not only for easy implementation of popular scalable QEP techniques,
-but often also for significantly improved utilization of GPU computing compared to solvers based on the Cholesky decomposition.
+Different from GPyTorch for Gaussian process (GP) models, Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch focuses on QEP, which generalizes GP by allowing flexible regularization on function spaces through a parameter $q>0$ and embraces GP as a special case with $q=2$. QEP is proven to be superior than GP in modeling inhomogeneous objects with abrupt changes or sharp contrast for $q<2$ [[Li et al (2023)]](https://papers.nips.cc/paper_files/paper/2023/hash/e6bfdd58f1326ff821a1b92743963bdf-Abstract-Conference.html).
+Inherited from GPyTorch, Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch has an efficient and scalable implementation by taking advantage of numerical linear algebra library [LinearOperator](https://github.com/cornellius-gp/linear_operator) and improved GPU utilization.
 
-Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch provides (1) significant GPU acceleration (through MVM based inference);
-(2) state-of-the-art implementations of the latest algorithmic advances for scalability and flexibility ([SKI/KISS-GP](http://proceedings.mlr.press/v37/wilson15.pdf), [stochastic Lanczos expansions](https://arxiv.org/abs/1711.03481), [LOVE](https://arxiv.org/pdf/1803.06058.pdf), [SKIP](https://arxiv.org/pdf/1802.08903.pdf), [stochastic variational](https://arxiv.org/pdf/1611.00336.pdf) [deep kernel learning](http://proceedings.mlr.press/v51/wilson16.pdf), ...);
-(3) easy integration with deep learning frameworks.
 
+<!--
+Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch features ...
+-->
 
-## Examples, Tutorials, and Documentation
 
-See our [**documentation, examples, tutorials**](https://qepytorch.readthedocs.io/en/stable/) on how to construct all sorts of models in Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch.
+## Tutorials, Examples, and Documentation
+
+See [**documentation**](https://qepytorch.readthedocs.io/en/stable/) on how to construct various QEP models in Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch.
 
 ## Installation
 
 **Requirements**:
 - Python >= 3.10
-- PyTorch >= 2.2
+- PyTorch >= 2.0
 - GPyTorch >= 1.13
 
+#### Stable Version
+
 Install Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch using pip or conda:
 
 ```bash
 pip install qpytorch
-conda install qpytorch -c qpytorch
+conda install qpytorch
 ```
 
 (To use packages globally but install Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch as a user-only package, use `pip install --user` above.)
 
-#### Latest (Unstable) Version
+#### Latest Version
 
-To upgrade to the latest (unstable) version, run
+To upgrade to the latest version, run
 
 ```bash
-pip install --upgrade git+https://github.com/cornellius-gp/linear_operator.git
-pip install --upgrade git+https://github.com/cornellius-gp/gpytorch.git
 pip install --upgrade git+https://github.com/lanzithinking/qepytorch.git
 ```
 
-#### Development version
+#### from source (for development)
 
 If you are contributing a pull request, it is best to perform a manual installation:
 
 ```sh
-git clone https://github.com/lanzithinking/qepytorch.git qpytorch
-cd qpytorch
+git clone https://github.com/lanzithinking/qepytorch.git
+cd qepytorch
+# either
 pip install -e .[dev,docs,examples,keops,pyro,test]  # keops and pyro are optional
+# or
+conda env create -f env_install.yaml # installed in the environment qpytorch
 ```
 
 <!--
@@ -102,7 +103,7 @@ To discuss any issues related to this AUR package refer to the comments section
 
 ## Citing Us
 
-If you use Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch, please cite the following papers:
+If you use Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch, please cite the following paper:
 > [Li, Shuyi, Michael O'Connor, and Shiwei Lan. "Bayesian Learning via Q-Exponential Process." In Advances in Neural Information Processing Systems (2023).](https://papers.nips.cc/paper_files/paper/2023/hash/e6bfdd58f1326ff821a1b92743963bdf-Abstract-Conference.html)
 ```
 @inproceedings{li2023QEP,
@@ -124,18 +125,18 @@ for information on submitting issues and pull requests.
 Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is primarily maintained by:
 - [Shiwei Lan](https://math.la.asu.edu/~slan) (Arizona State University)
 
-We would like to thank our other contributors including (but not limited to)
-Shuyi Li,
+Thanks to the following contributors including (but not limited to)
+- Shuyi Li,
 Guangting Yu,
 Zhi Chang,
 Chukwudi Paul Obite,
 Keyan Wu,
 and many more!
 
-
+<!--
 ## Acknowledgements
 Development of Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is supported by.
-
+-->
 
 ## License
 

{qpytorch-0.1 → qpytorch-0.1.1}/README.md

@@ -6,60 +6,61 @@
 [![License](https://img.shields.io/badge/license-MIT-green.svg)](LICENSE)
 
 [![Python Version](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/)
-[![Conda](https://img.shields.io/conda/v/conda-forge/qpytorch.svg)](https://anaconda.org/conda-forge/qpytorch)
+[![Conda](https://img.shields.io/conda/v/qpytorch.svg)](https://anaconda.org/qpytorch)
 [![PyPI](https://img.shields.io/pypi/v/qpytorch.svg)](https://pypi.org/project/qpytorch)
 
-Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is a Q-exponential process library implemented using PyTorch built on [GPyTorch](https://gpytorch.ai). Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is designed for creating scalable, flexible, and modular Q-exponential process models with ease.
+Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is a Python package for Q-exponential process ([QEP](https://papers.nips.cc/paper_files/paper/2023/file/e6bfdd58f1326ff821a1b92743963bdf-Paper-Conference.pdf)) implemented using PyTorch and built up on [GPyTorch](https://gpytorch.ai). Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is designed to facilitate creating scalable, flexible, and modular QPE models.
 
-Internally, Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch differs from many existing approaches to QEP inference by performing most inference operations using numerical linear algebra techniques like preconditioned conjugate gradients.
-Implementing a scalable QEP method is as simple as providing a matrix multiplication routine with the kernel matrix and its derivative via our [LinearOperator](https://github.com/cornellius-gp/linear_operator) interface,
-or by composing many of our already existing `LinearOperators`.
-This allows not only for easy implementation of popular scalable QEP techniques,
-but often also for significantly improved utilization of GPU computing compared to solvers based on the Cholesky decomposition.
+Different from GPyTorch for Gaussian process (GP) models, Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch focuses on QEP, which generalizes GP by allowing flexible regularization on function spaces through a parameter $q>0$ and embraces GP as a special case with $q=2$. QEP is proven to be superior than GP in modeling inhomogeneous objects with abrupt changes or sharp contrast for $q<2$ [[Li et al (2023)]](https://papers.nips.cc/paper_files/paper/2023/hash/e6bfdd58f1326ff821a1b92743963bdf-Abstract-Conference.html).
+Inherited from GPyTorch, Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch has an efficient and scalable implementation by taking advantage of numerical linear algebra library [LinearOperator](https://github.com/cornellius-gp/linear_operator) and improved GPU utilization.
 
-Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch provides (1) significant GPU acceleration (through MVM based inference);
-(2) state-of-the-art implementations of the latest algorithmic advances for scalability and flexibility ([SKI/KISS-GP](http://proceedings.mlr.press/v37/wilson15.pdf), [stochastic Lanczos expansions](https://arxiv.org/abs/1711.03481), [LOVE](https://arxiv.org/pdf/1803.06058.pdf), [SKIP](https://arxiv.org/pdf/1802.08903.pdf), [stochastic variational](https://arxiv.org/pdf/1611.00336.pdf) [deep kernel learning](http://proceedings.mlr.press/v51/wilson16.pdf), ...);
-(3) easy integration with deep learning frameworks.
 
+<!--
+Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch features ...
+-->
 
-## Examples, Tutorials, and Documentation
 
-See our [**documentation, examples, tutorials**](https://qepytorch.readthedocs.io/en/stable/) on how to construct all sorts of models in Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch.
+## Tutorials, Examples, and Documentation
+
+See [**documentation**](https://qepytorch.readthedocs.io/en/stable/) on how to construct various QEP models in Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch.
 
 ## Installation
 
 **Requirements**:
 - Python >= 3.10
-- PyTorch >= 2.2
+- PyTorch >= 2.0
 - GPyTorch >= 1.13
 
+#### Stable Version
+
 Install Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch using pip or conda:
 
 ```bash
 pip install qpytorch
-conda install qpytorch -c qpytorch
+conda install qpytorch
 ```
 
 (To use packages globally but install Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch as a user-only package, use `pip install --user` above.)
 
-#### Latest (Unstable) Version
+#### Latest Version
 
-To upgrade to the latest (unstable) version, run
+To upgrade to the latest version, run
 
 ```bash
-pip install --upgrade git+https://github.com/cornellius-gp/linear_operator.git
-pip install --upgrade git+https://github.com/cornellius-gp/gpytorch.git
 pip install --upgrade git+https://github.com/lanzithinking/qepytorch.git
 ```
 
-#### Development version
+#### from source (for development)
 
 If you are contributing a pull request, it is best to perform a manual installation:
 
 ```sh
-git clone https://github.com/lanzithinking/qepytorch.git qpytorch
-cd qpytorch
+git clone https://github.com/lanzithinking/qepytorch.git
+cd qepytorch
+# either
 pip install -e .[dev,docs,examples,keops,pyro,test]  # keops and pyro are optional
+# or
+conda env create -f env_install.yaml # installed in the environment qpytorch
 ```
 
 <!--
@@ -80,7 +81,7 @@ To discuss any issues related to this AUR package refer to the comments section
 
 ## Citing Us
 
-If you use Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch, please cite the following papers:
+If you use Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch, please cite the following paper:
 > [Li, Shuyi, Michael O'Connor, and Shiwei Lan. "Bayesian Learning via Q-Exponential Process." In Advances in Neural Information Processing Systems (2023).](https://papers.nips.cc/paper_files/paper/2023/hash/e6bfdd58f1326ff821a1b92743963bdf-Abstract-Conference.html)
 ```
 @inproceedings{li2023QEP,
@@ -102,18 +103,18 @@ for information on submitting issues and pull requests.
 Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is primarily maintained by:
 - [Shiwei Lan](https://math.la.asu.edu/~slan) (Arizona State University)
 
-We would like to thank our other contributors including (but not limited to)
-Shuyi Li,
+Thanks to the following contributors including (but not limited to)
+- Shuyi Li,
 Guangting Yu,
 Zhi Chang,
 Chukwudi Paul Obite,
 Keyan Wu,
 and many more!
 
-
+<!--
 ## Acknowledgements
 Development of Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is supported by.
-
+-->
 
 ## License
 

qpytorch-0.1.1/docs/_config.yml

@@ -0,0 +1,6 @@
+title: QePyTorch
+description: Q-Exponential Process Library built on PyTorch and GPyTorch
+show_downloads: true
+remote_theme: pages-themes/leap-day@v0.2.0
+plugins:
+- jekyll-remote-theme # add this line to the plugins list if you already have one

qpytorch-0.1.1/docs/index.md

@@ -0,0 +1,59 @@
+---
+layout: default
+---
+
+*A highly efficient implementation of Q-Exponential Processes in [PyTorch](https://pytorch.org) based on [GPyTorch](https://gpytorch.ai).*
+
+* * *
+
+# Example
+
+Check out the [`examples`](https://github.com/lanzithinking/QePyTorch/tree/main/examples).
+
+# Documentation
+
+Here is the [`documentation`](https://qepytorch.readthedocs.io/en/stable/).
+
+* * *
+
+# Installation
+
+## pip
+
+```
+pip install qpytorch
+```
+
+## conda
+
+```
+conda install -c conda-forge qpytorch
+```
+
+## from source
+
+```
+git clone https://github.com/lanzithinking/qepytorch.git
+cd qepytorch
+conda env create -f env_install.yaml
+```
+
+* * *
+
+# Team
+
+* Maintain: [Shiwei Lan](https://math.la.asu.edu/~slan) (Arizona State University)
+* Contribute: Shuyi Li, Guangting Yu, Zhi Chang, Chukwudi Paul Obite, Keyan Wu, and many more!
+
+# Citing Us
+
+> [Li, Shuyi, Michael O'Connor, and Shiwei Lan. "Bayesian Learning via Q-Exponential Process." In Advances in Neural Information Processing Systems (2023).](https://papers.nips.cc/paper_files/paper/2023/hash/e6bfdd58f1326ff821a1b92743963bdf-Abstract-Conference.html)
+
+```
+@inproceedings{li2023QEP,
+  title={Bayesian Learning via Q-Exponential Process},
+  author={Li, Shuyi, Michael O'Connor, and Shiwei Lan},
+  booktitle={Advances in Neural Information Processing Systems},
+  year={2023}
+}
+```

qpytorch-0.1/venv_install.yaml → qpytorch-0.1.1/env_install.yaml

@@ -7,6 +7,5 @@ dependencies:
   - pip
   - pip:
       - -e .[dev,docs,examples,keops,pyro,test]
-      - numpy<2.0 # Numpy 2.0 is not fully supported
 
-# conda env create -f venv_install.yaml
+# conda env create -f env_install.yaml

{qpytorch-0.1 → qpytorch-0.1.1}/qpytorch/version.py

@@ -1,4 +1,4 @@
 # file generated by setuptools_scm
 # don't change, don't track in version control
-__version__ = version = '0.1'
-__version_tuple__ = version_tuple = (0, 1)
+__version__ = version = '0.1.1'
+__version_tuple__ = version_tuple = (0, 1, 1)

{qpytorch-0.1 → qpytorch-0.1.1}/qpytorch.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: qpytorch
-Version: 0.1
+Version: 0.1.1
 Summary: An implementation of Q-Exponential Processes in Pytorch based on GPyTorch
 Home-page: https://lanzithinking.github.io/qepytorch/
 Author: Shiwei Lan
@@ -28,60 +28,61 @@ License-File: LICENSE
 [![License](https://img.shields.io/badge/license-MIT-green.svg)](LICENSE)
 
 [![Python Version](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/)
-[![Conda](https://img.shields.io/conda/v/conda-forge/qpytorch.svg)](https://anaconda.org/conda-forge/qpytorch)
+[![Conda](https://img.shields.io/conda/v/qpytorch.svg)](https://anaconda.org/qpytorch)
 [![PyPI](https://img.shields.io/pypi/v/qpytorch.svg)](https://pypi.org/project/qpytorch)
 
-Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is a Q-exponential process library implemented using PyTorch built on [GPyTorch](https://gpytorch.ai). Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is designed for creating scalable, flexible, and modular Q-exponential process models with ease.
+Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is a Python package for Q-exponential process ([QEP](https://papers.nips.cc/paper_files/paper/2023/file/e6bfdd58f1326ff821a1b92743963bdf-Paper-Conference.pdf)) implemented using PyTorch and built up on [GPyTorch](https://gpytorch.ai). Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is designed to facilitate creating scalable, flexible, and modular QPE models.
 
-Internally, Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch differs from many existing approaches to QEP inference by performing most inference operations using numerical linear algebra techniques like preconditioned conjugate gradients.
-Implementing a scalable QEP method is as simple as providing a matrix multiplication routine with the kernel matrix and its derivative via our [LinearOperator](https://github.com/cornellius-gp/linear_operator) interface,
-or by composing many of our already existing `LinearOperators`.
-This allows not only for easy implementation of popular scalable QEP techniques,
-but often also for significantly improved utilization of GPU computing compared to solvers based on the Cholesky decomposition.
+Different from GPyTorch for Gaussian process (GP) models, Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch focuses on QEP, which generalizes GP by allowing flexible regularization on function spaces through a parameter $q>0$ and embraces GP as a special case with $q=2$. QEP is proven to be superior than GP in modeling inhomogeneous objects with abrupt changes or sharp contrast for $q<2$ [[Li et al (2023)]](https://papers.nips.cc/paper_files/paper/2023/hash/e6bfdd58f1326ff821a1b92743963bdf-Abstract-Conference.html).
+Inherited from GPyTorch, Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch has an efficient and scalable implementation by taking advantage of numerical linear algebra library [LinearOperator](https://github.com/cornellius-gp/linear_operator) and improved GPU utilization.
 
-Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch provides (1) significant GPU acceleration (through MVM based inference);
-(2) state-of-the-art implementations of the latest algorithmic advances for scalability and flexibility ([SKI/KISS-GP](http://proceedings.mlr.press/v37/wilson15.pdf), [stochastic Lanczos expansions](https://arxiv.org/abs/1711.03481), [LOVE](https://arxiv.org/pdf/1803.06058.pdf), [SKIP](https://arxiv.org/pdf/1802.08903.pdf), [stochastic variational](https://arxiv.org/pdf/1611.00336.pdf) [deep kernel learning](http://proceedings.mlr.press/v51/wilson16.pdf), ...);
-(3) easy integration with deep learning frameworks.
 
+<!--
+Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch features ...
+-->
 
-## Examples, Tutorials, and Documentation
 
-See our [**documentation, examples, tutorials**](https://qepytorch.readthedocs.io/en/stable/) on how to construct all sorts of models in Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch.
+## Tutorials, Examples, and Documentation
+
+See [**documentation**](https://qepytorch.readthedocs.io/en/stable/) on how to construct various QEP models in Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch.
 
 ## Installation
 
 **Requirements**:
 - Python >= 3.10
-- PyTorch >= 2.2
+- PyTorch >= 2.0
 - GPyTorch >= 1.13
 
+#### Stable Version
+
 Install Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch using pip or conda:
 
 ```bash
 pip install qpytorch
-conda install qpytorch -c qpytorch
+conda install qpytorch
 ```
 
 (To use packages globally but install Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch as a user-only package, use `pip install --user` above.)
 
-#### Latest (Unstable) Version
+#### Latest Version
 
-To upgrade to the latest (unstable) version, run
+To upgrade to the latest version, run
 
 ```bash
-pip install --upgrade git+https://github.com/cornellius-gp/linear_operator.git
-pip install --upgrade git+https://github.com/cornellius-gp/gpytorch.git
 pip install --upgrade git+https://github.com/lanzithinking/qepytorch.git
 ```
 
-#### Development version
+#### from source (for development)
 
 If you are contributing a pull request, it is best to perform a manual installation:
 
 ```sh
-git clone https://github.com/lanzithinking/qepytorch.git qpytorch
-cd qpytorch
+git clone https://github.com/lanzithinking/qepytorch.git
+cd qepytorch
+# either
 pip install -e .[dev,docs,examples,keops,pyro,test]  # keops and pyro are optional
+# or
+conda env create -f env_install.yaml # installed in the environment qpytorch
 ```
 
 <!--
@@ -102,7 +103,7 @@ To discuss any issues related to this AUR package refer to the comments section
 
 ## Citing Us
 
-If you use Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch, please cite the following papers:
+If you use Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch, please cite the following paper:
 > [Li, Shuyi, Michael O'Connor, and Shiwei Lan. "Bayesian Learning via Q-Exponential Process." In Advances in Neural Information Processing Systems (2023).](https://papers.nips.cc/paper_files/paper/2023/hash/e6bfdd58f1326ff821a1b92743963bdf-Abstract-Conference.html)
 ```
 @inproceedings{li2023QEP,
@@ -124,18 +125,18 @@ for information on submitting issues and pull requests.
 Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is primarily maintained by:
 - [Shiwei Lan](https://math.la.asu.edu/~slan) (Arizona State University)
 
-We would like to thank our other contributors including (but not limited to)
-Shuyi Li,
+Thanks to the following contributors including (but not limited to)
+- Shuyi Li,
 Guangting Yu,
 Zhi Chang,
 Chukwudi Paul Obite,
 Keyan Wu,
 and many more!
 
-
+<!--
 ## Acknowledgements
 Development of Q<sup style="font-size: 0.5em;">&#9428;</sup>PyTorch is supported by.
-
+-->
 
 ## License
 

{qpytorch-0.1 → qpytorch-0.1.1}/qpytorch.egg-info/SOURCES.txt

@@ -3,14 +3,16 @@
 CONTRIBUTING.md
 LICENSE
 README.md
+env_install.yaml
 pyproject.toml
 setup.cfg
 setup.py
-venv_install.yaml
 .conda/meta.yaml
 .github/workflows/deploy.yml
 .github/workflows/run_test_suite.yml
 docs/Makefile
+docs/_config.yml
+docs/index.md
 docs/source/.gitignore
 docs/source/beta_features.rst
 docs/source/conf.py

{qpytorch-0.1 → qpytorch-0.1.1}/qpytorch.egg-info/requires.txt

@@ -4,6 +4,7 @@ scikit-learn
 scipy>=1.6.0
 linear_operator>=0.6
 gpytorch>=1.13
+numpy<2
 
 [dev]
 pre-commit

{qpytorch-0.1 → qpytorch-0.1.1}/setup.py

@@ -37,7 +37,7 @@ def find_version(*file_paths):
 readme = open("README.md").read()
 
 
-torch_min = "2.2"
+torch_min = "2.0"
 install_requires = [
     "jaxtyping",
     "mpmath>=0.19,<=1.3",  # avoid incompatibiltiy with torch+sympy with mpmath 1.4
@@ -45,6 +45,7 @@ install_requires = [
     "scipy>=1.6.0",
     "linear_operator>=0.6",
     "gpytorch>=1.13",
+    "numpy<2",  # avoid incompatibility of torch with numpy 2 (the minimum supporting (numpy 2) version of pytorch is 2.3)
 ]
 # if recent dev version of PyTorch is installed, no need to install stable
 try: