purin 1.0.0__tar.gz

purin-1.0.0/AUTHORS

@@ -0,0 +1,4 @@
+SOULS-TAMU contributors
+
+Upstream research authors include Ashfaq Iftakher, Rahul Golder,
+Bimol Nath Roy, and M. M. Faruque Hasan. See CITATION.bib and NOTICE.

purin-1.0.0/CHANGELOG.md

@@ -0,0 +1,43 @@
+# Changelog
+
+Notable changes to PURIN are recorded here. This history starts with the
+initial public version; internal development steps are not separate releases.
+The first GitHub publication contains the initial source distribution.
+Version 1.0.0 remains unreleased until a release is explicitly published;
+no release date is assigned.
+
+## 1.0.0 — Unreleased
+
+- Promote the shared-stack implementation to `main`; preserve the previous main in `v0.1.0`.
+- Align main-branch installation links, badges and citation metadata.
+
+### Added
+
+- Added dense-reference KKT/Pallas-interpreter checks, barrier derivatives,
+  sparse adjoint validation and full-state projection pullback tests.
+
+- API, numerical, frontend integration, dependency review, documentation,
+  quality regression and informational performance workflows.
+
+- Expanded analytic, dense-reference, sparse-kernel and implicit-gradient tests,
+  with an inclusive 90% combined statement-and-branch coverage gate.
+
+- Cross-platform CPU CI for Linux, Windows and Apple Silicon macOS, with
+  baseline/current JAX, runtime configuration artifacts and ARM64 coverage.
+
+- GitHub Actions CI for numerical tests, package checks and Ruff, with measured
+  branch coverage reporting and scheduled Python CodeQL analysis.
+
+- A standalone JAX library for constrained projections and sparse numerical
+  kernels, usable independently of OHAGI and the frontend frameworks.
+- Projection problem interfaces and reusable preparation utilities.
+- Euclidean distance, diagonal weighted-distance, local quadratic and
+  sequential projection interfaces.
+- Linear equality, Chambolle–Pock, ADMM, Fischer–Burmeister and primal-dual
+  interior point projection methods.
+- Sparse operators and projection differentiation utilities, including
+  implicit differentiation support.
+- Shared projection infrastructure for OHAGI, KKTHN, DAEHN and NLPOPTNet.
+- Python packaging, a runnable quick start, contribution guidance and a
+  BibTeX software citation.
+- Apache-2.0 licensing and attribution notices.

purin-1.0.0/CITATION.bib

@@ -0,0 +1,8 @@
+@misc{purin,
+  author       = {{SOULS-TAMU contributors}},
+  title        = {{PURIN}: Differentiable constrained projections and sparse numerical kernels in {JAX}},
+  howpublished = {Computer software},
+  version      = {1.0.0},
+  url          = {https://github.com/SOULS-TAMU/Purin},
+  note         = {Version 1.0.0, initial release in preparation. Licensed under Apache-2.0}
+}

purin-1.0.0/CITATION.cff

@@ -0,0 +1,10 @@
+cff-version: 1.2.0
+message: "If you use PURIN, please cite this software. A BibTeX entry is available in CITATION.bib."
+type: software
+title: "PURIN: Differentiable constrained projections and sparse numerical kernels in JAX"
+authors:
+  - name: "SOULS-TAMU contributors"
+version: "1.0.0"
+url: "https://github.com/SOULS-TAMU/Purin"
+repository-code: "https://github.com/SOULS-TAMU/Purin"
+license: Apache-2.0

purin-1.0.0/CONTRIBUTING.md

@@ -0,0 +1,164 @@
+# Contributing to PURIN
+
+Thank you for helping improve PURIN. Bug reports, documentation fixes,
+examples and numerical improvements are welcome. This guide accompanies
+the first public version of the project.
+
+## Issues and proposals
+
+For small fixes, open a focused
+[pull request](https://github.com/SOULS-TAMU/Purin/pulls). Discuss new solvers,
+public APIs, dependencies or changes to numerical behavior in an
+[issue](https://github.com/SOULS-TAMU/Purin/issues) first.
+
+For a bug report, provide a minimal reproducer, expected and observed
+behavior, package versions, device, dtype, random seed and solver settings.
+Include the problem dimensions and constraint residuals when relevant.
+Remove private data and credentials from examples and logs.
+
+## Development setup
+
+Use Python 3.10 or newer in an isolated environment. Fork the repository
+on GitHub for contributions, then clone your fork. For a local checkout of
+the upstream source:
+
+```sh
+git clone https://github.com/SOULS-TAMU/Purin.git
+cd Purin
+python -m pip install -e ".[build]"
+python -m pip install pytest ruff
+```
+
+Create a focused branch for your changes and submit the pull request against
+`main`.
+
+Follow the [JAX installation guide](https://docs.jax.dev/en/latest/installation.html)
+for your device. Install OHAGI and the relevant frontend in editable mode
+only when checking their integration. PURIN itself must remain usable
+without those packages.
+
+## Library boundaries
+
+PURIN owns projection problems, composition, solvers, sparse operators and
+projection differentiation. Reuse its canonical solver implementations.
+Keep training, neural network construction, checkpointing and model export
+in OHAGI; keep framework-specific formulation policies in the frontends.
+
+Do not introduce imports from OHAGI, KKTHN, DAEHN or NLPOPTNet into PURIN.
+Avoid duplicating numerical kernels to support a new adapter.
+
+## Code and documentation
+
+Use descriptive names, focused functions and Google-style docstrings for
+public interfaces. Document shapes, axis order, coordinate conventions,
+constraint signs, metric assumptions and gradient behavior where relevant.
+Explain mathematical choices and invariants rather than restating Python.
+
+Run these checks from the PURIN directory:
+
+```sh
+python -m ruff format --check src
+python -m ruff check --select F,E9 src
+```
+
+Use `python -m ruff format src` to apply formatting. Keep unrelated
+formatting changes separate and leave benchmark scripts unchanged unless
+the proposed change specifically concerns them.
+
+## Numerical validation
+
+Include focused regression tests or a reproducible validation script with
+the pull request. Record commands and results. Run the README quick start
+when changing the public projection workflow.
+
+| Change | Validation |
+| --- | --- |
+| Solver or projection | Compare projected values, feasibility residuals, objective values and available convergence diagnostics. |
+| Differentiation | Compare parameter and input gradients with the previous implementation; use an independent check on suitable smooth problems. |
+| Sparse kernels | Compare with an equivalent small dense calculation and exercise supported shapes and dtypes. |
+| Public interfaces | Check signatures, configuration defaults and the relevant unchanged frontend benchmarks. |
+| Restructuring | Compare outputs and gradients before and after the change under identical settings. |
+
+Use identical inputs, seeds, devices, JAX versions, precision and solver
+settings for comparisons. State tolerances explicitly and investigate any
+difference. Cover applicable boundary cases, including empty constraint
+sets, degenerate systems, iteration limits and warm starts. At nonsmooth
+points, check the documented differentiation convention rather than
+assuming a unique derivative.
+
+Separate intentional numerical changes from restructuring. For timing
+comparisons, warm up compilation and synchronize device execution.
+
+## Packaging
+
+From the PURIN directory:
+
+```sh
+python -m build
+python -m twine check dist/*
+```
+
+Install the built wheel in a separate environment and run the quick start.
+Keep environments, caches, build output, generated results and model
+artifacts out of version control.
+
+## Pull request checklist
+
+- Explain the problem, resulting behavior and compatibility implications.
+- Keep changes focused and preserve public APIs and benchmark call signatures.
+- Include reproducible numerical checks and explicit tolerances.
+- Update documentation and the appropriate unreleased changelog entry.
+- Preserve the dependency boundary between PURIN and its consumers.
+
+Contributions are provided under the project's [Apache-2.0 license](LICENSE).
+Retain relevant attribution in [AUTHORS](AUTHORS) and [NOTICE](NOTICE).
+Keep [CITATION.bib](CITATION.bib) and [CITATION.cff](CITATION.cff) consistent.
+The CFF metadata supports GitHub's citation panel; BibTeX remains available directly.
+
+## Automated checks
+
+CI runs the full numerical suite on CPU across these configurations:
+
+| Platform | Python | JAX configuration |
+| --- | --- | --- |
+| Linux x86-64, Windows x86-64, macOS Apple Silicon | 3.10–3.14 | Newest compatible dependencies for each interpreter. |
+| Linux x86-64, Windows x86-64, macOS Apple Silicon | 3.12 | JAX/JAXlib 0.6.2 baseline. |
+| Linux ARM64 | 3.13 | Newest compatible dependencies. |
+
+The configuration tests additionally exercise float32/float64, eager/JIT
+execution and batch sizes one and four for the equality projection and
+functional backbone interfaces. Full solver regression tests use float64.
+Ruff runs on Linux; wheel and source-distribution builds and installation
+checks run on Linux, Windows and macOS. Each core test job uploads its OS,
+architecture, Python, package versions and JAX devices as a diagnostic artifact.
+The test suite is in `tests/`. CPU results do not establish GPU/TPU support.
+
+Install the test dependencies with `python -m pip install ".[test]"`, then run:
+
+```sh
+python -m pytest -q --cov=purin --cov-branch --cov-report=term-missing --cov-report=xml --cov-report=json
+```
+
+Coverage measures the entire package, including unexecuted modules. The badge
+combines statement and branch coverage from the Linux x86-64 / Python 3.12 / current-JAX core test run.
+Reports are retained as workflow artifacts; the badge and measured totals are
+published to `codex/coverage` after successful trusted pushes to `main`.
+Pull requests run checks with read-only repository access and do not publish
+coverage. No third-party coverage account or additional secret is required.
+
+CodeQL runs extended Python security analysis on pushes, pull requests and a
+weekly schedule. Its badge reports workflow execution status; inspect code
+scanning alerts separately. GitHub Actions dependencies are pinned to commits
+and monitored by Dependabot.
+
+The combined statement-and-branch coverage floor is 90%. Every source module,
+including the dense reference solvers, remains in the denominator. Do not omit
+modules to improve the score.
+
+## Compatibility and quality regression checks
+
+Additional workflows protect public APIs, numerical results and frontend
+integration, validate documentation and dependency changes, and publish code
+quality and performance reports. See the [validation policies](.github/validation/README.md)
+for thresholds, existing baselines and reproduction instructions. Timing changes
+are informational; correctness and quality regressions fail their checks.

purin-1.0.0/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2026 MM Faruque Hasan, Bimol Nath Roy and Rahul Golder
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

purin-1.0.0/MANIFEST.in

@@ -0,0 +1,4 @@
+include AUTHORS CHANGELOG.md CITATION.bib CITATION.cff CONTRIBUTING.md NOTICE
+
+recursive-include docs *.rst *.md *.py *.css *.html *.json *.txt *.png
+recursive-include assets *.png *.svg

purin-1.0.0/NOTICE

@@ -0,0 +1,12 @@
+PURIN
+
+Copyright (c) 2025 SOULS-TAMU
+
+Part of the HardNet stack maintained from SOULS-TAMU implementations.
+KKT-HardNet, DAEHN, and NLPOpt-Net contributions retain their
+upstream authorship and applicable license terms.
+Original repositories: https://github.com/SOULS-TAMU/kkt-hardnet
+and https://github.com/SOULS-TAMU/nlpoptnet.
+
+The current project is licensed under the Apache License, Version 2.0.
+See LICENSE for the full terms.

purin-1.0.0/PKG-INFO

@@ -0,0 +1,205 @@
+Metadata-Version: 2.4
+Name: purin
+Version: 1.0.0
+Summary: Differentiable constrained projections and sparse numerical kernels in JAX
+Author: SOULS-TAMU contributors
+License-Expression: Apache-2.0
+Project-URL: Documentation, https://purin.readthedocs.io/en/latest/
+Project-URL: Homepage, https://github.com/SOULS-TAMU/Purin
+Project-URL: Repository, https://github.com/SOULS-TAMU/Purin
+Project-URL: Issues, https://github.com/SOULS-TAMU/Purin/issues
+Project-URL: Changelog, https://github.com/SOULS-TAMU/Purin/blob/main/CHANGELOG.md
+Keywords: jax,constrained-learning,differentiable-optimization
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Programming Language :: Python :: 3.15
+Classifier: Operating System :: OS Independent
+Classifier: Intended Audience :: Science/Research
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+License-File: NOTICE
+Requires-Dist: jax>=0.4.30
+Requires-Dist: numpy>=1.26
+Requires-Dist: scipy>=1.11
+Provides-Extra: docs
+Requires-Dist: Sphinx<9,>=8.2; extra == "docs"
+Requires-Dist: sphinx-rtd-theme<4,>=3.1; extra == "docs"
+Requires-Dist: myst-parser<5,>=4; extra == "docs"
+Provides-Extra: test
+Requires-Dist: pytest>=8; extra == "test"
+Requires-Dist: pytest-cov>=5; extra == "test"
+Requires-Dist: osqp>=1; extra == "test"
+Provides-Extra: build
+Requires-Dist: build>=1; extra == "build"
+Requires-Dist: twine>=5; extra == "build"
+Provides-Extra: cuda12
+Requires-Dist: jax[cuda12]>=0.4.30; extra == "cuda12"
+Dynamic: license-file
+
+<p align="center">
+  <img src="https://raw.githubusercontent.com/SOULS-TAMU/Purin/main/assets/purin-logo-v2.png" alt="PURIN logo" width="200">
+</p>
+
+# PURIN
+
+**Differentiable constraint projections and sparse numerical kernels in JAX.**
+
+[![CI](https://github.com/SOULS-TAMU/Purin/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/SOULS-TAMU/Purin/actions/workflows/ci.yml?query=branch%3Amain)
+[![Validation](https://github.com/SOULS-TAMU/Purin/actions/workflows/validation.yml/badge.svg?branch=main)](https://github.com/SOULS-TAMU/Purin/actions/workflows/validation.yml?query=branch%3Amain)
+[![CodeQL](https://github.com/SOULS-TAMU/Purin/actions/workflows/codeql.yml/badge.svg?branch=main)](https://github.com/SOULS-TAMU/Purin/actions/workflows/codeql.yml?query=branch%3Amain)
+[![Dependency review](https://github.com/SOULS-TAMU/Purin/actions/workflows/dependency-review.yml/badge.svg?branch=main)](https://github.com/SOULS-TAMU/Purin/actions/workflows/dependency-review.yml?query=branch%3Amain)
+[![Coverage](https://raw.githubusercontent.com/SOULS-TAMU/Purin/codex/coverage/coverage.svg)](https://github.com/SOULS-TAMU/Purin/tree/codex/coverage)
+
+[![Python 3.10–3.15](https://img.shields.io/badge/Python-3.10%E2%80%933.15-3776AB?logo=python&logoColor=white)](https://github.com/SOULS-TAMU/Purin/blob/main/.github/workflows/release.yml)
+[![CPU platforms](https://img.shields.io/badge/CPU-Linux%20%7C%20Windows%20%7C%20macOS-555555)](https://github.com/SOULS-TAMU/Purin/blob/main/.github/workflows/ci.yml)
+[![JAX](https://img.shields.io/badge/Powered_by-JAX-8A2BE2)](https://docs.jax.dev/)
+[![License: Apache 2.0](https://img.shields.io/badge/License-Apache_2.0-blue)](https://github.com/SOULS-TAMU/Purin/blob/main/LICENSE)
+[![Citation: BibTeX](https://img.shields.io/badge/Cite-BibTeX-6F4E37)](https://github.com/SOULS-TAMU/Purin/blob/main/CITATION.bib)
+
+Turn a parameterized constraint problem into a reusable projection layer.
+Use PURIN independently or as the numerical backend for OHAGI and the HardNet
+frontends.
+
+- **Projection formulations:** Euclidean distance, diagonal weighted distance
+  and objective-aware local quadratic corrections.
+- **Six solver methods:** affine equality projection, Chambolle–Pock, ADMM,
+  full/reduced Fischer–Burmeister and primal-dual interior point.
+- **Sparse computation:** reusable operators and kernels, including matrix-free
+  paths where supported by the selected solver.
+- **Differentiable layers:** solver-specific implicit differentiation and
+  numerical diagnostics for constrained learning.
+
+[Installation](#installation) · [Basic API](#basic-api) · [Projection methods](#projection-methods) · [Documentation](#documentation) · [Contributing](#contributing) · [Citation](#citation)
+
+## Installation
+
+Requires **Python 3.10 or newer**. Install from PyPI:
+
+```shell
+python -m pip install purin
+```
+
+From a local stack checkout, use `python -m pip install ./purin`.
+PURIN does not depend on OHAGI or any frontend package.
+
+## Basic API
+
+Project a point onto the affine equality **u + v = b**:
+
+```python
+import jax
+import jax.numpy as jnp
+from purin import Distance, Projection
+
+jax.config.update("jax_enable_x64", True)
+
+problem = Projection()
+p = problem.add_parameter(["b"])
+y = problem.add_variable(["u", "v"])
+problem.objective = 0.0
+problem.constraints.add(y.u + y.v == p.b)
+
+project = Distance(problem, "linear-equality").prepare(batch_size=1)
+candidate = jnp.array([[2.0, 0.0]])  # Shape: (batch, variables).
+parameters = jnp.array([[1.0]])     # Shape: (batch, parameters).
+projected = project(candidate, parameters)
+print(projected)  # [[ 1.5 -0.5]]
+```
+
+Prepare once and reuse the layer for its configured shapes and dtype. The same
+problem interface accepts other registered solver methods when their constraint
+requirements are satisfied.
+
+## Projection methods
+
+Choose a solver for the constraint structure of your projection problem:
+
+| Algorithm / method name | Suitable problems | Implementation notes |
+| --- | --- | --- |
+| Linear equality (`linear-equality`) | Euclidean projection onto affine equalities, such as conservation equations or linear balance constraints. | Direct equality-system solve. Requires verified affine constraints; inequalities and nonlinear equalities are rejected. |
+| Chambolle–Pock (`chambolle-pock`) | Convex distance projections with affine equalities, affine inequalities and bounds; useful for sparse polyhedral constraints. | First-order primal-dual splitting with fixed-step or accelerated updates through `ChambollePockConfig(accelerated=...)`. Both variants use shared sparse kernels. |
+| ADMM (`admm`) | Convex distance projections with affine equalities, affine inequalities and bounds, including structured linear constraints. | OSQP-style operator splitting with configurable penalty and linear solves. Requires verified affine constraints. |
+| Full Fischer–Burmeister Newton (`fischer-burmeister-newton`) | Distance projections with smooth nonlinear equalities and inequalities, as well as affine constraints. | Solves the full stationarity, feasibility and complementarity system using a Fischer–Burmeister reformulation. |
+| Reduced Fischer–Burmeister (`fischer-burmeister-reduced`) | The same affine or smooth nonlinear constrained projection class, especially when a reduced sparse Newton system is useful. | Uses pivot reduction and sparse linear-system operators for the complementarity equations. |
+| Primal-dual interior point (`primal-dual-interior-point`) | Distance projections with smooth equality and inequality constraints, including nonlinear formulations. | Uses primal-dual barrier iterations and Newton systems with positive slack and inequality-multiplier iterates. |
+
+`Distance` defines Euclidean projection; `WeightedDistance` supplies a positive
+diagonal metric for the supported CP, ADMM and interior-point interfaces.
+`LocalQuadratic` builds a positive diagonal quadratic objective and linearizes
+constraints around an anchor, then solves the local QP with CP. Its nonlinear
+constraints are approximated in that subproblem. `Sequential` composes
+corrections; it is not a separate numerical solver.
+
+Lower-level closed-form `BoxProjection` (bound clipping) and `AffineProjection`
+(affine equality projection) utilities are available in
+`purin.projections.closed_form`. Use `registered_methods()` to list the six
+solver method names above; these utilities are not additional registry entries.
+
+For nonlinear, nonconvex constraints, convergence to a local stationary point
+does not establish a globally nearest feasible point. Check solver diagnostics
+and residuals, and choose precision and tolerances for your problem.
+
+## The HardNet stack
+
+- **[KKTHN](https://github.com/SOULS-TAMU/kkt-hardnet/tree/main),
+  [DAEHN](https://github.com/SOULS-TAMU/dae-hardnet/tree/main) and
+  [NLPOPTNet](https://github.com/SOULS-TAMU/nlpoptnet)** supply problem formulations
+  and framework-specific processing.
+- **[OHAGI](https://github.com/SOULS-TAMU/Ohagi/tree/main)** composes models,
+  runs the shared learning engine, and handles checkpoints and JAX deployment.
+- **[PURIN](https://github.com/SOULS-TAMU/Purin/tree/main)** owns projection
+  problems, numerical solvers, sparse kernels and projection differentiation.
+
+## Compatibility and validation
+
+The PyPI CPU installation matrix covers **Python 3.10–3.15** on Linux, Windows and macOS,
+plus Linux ARM64 on Python 3.13. It uses current compatible JAX releases;
+regular CI additionally tests JAX 0.6.2 on Python 3.12. See the [release workflow](https://github.com/SOULS-TAMU/Purin/blob/main/.github/workflows/release.yml)
+for exact environments; accelerator installation follows the
+[JAX guide](https://docs.jax.dev/en/latest/installation.html).
+
+- **CI:** tests, distribution builds and installation checks.
+- **Validation:** public APIs, numerical references, documentation and code quality.
+- **Coverage:** at least **90% combined statement and branch coverage**; the badge
+  links to the measured report for this branch.
+- **Security:** CodeQL analysis and dependency review.
+
+## Documentation
+
+**[Read the full documentation](https://purin.readthedocs.io/en/latest/)** — installation, tutorials, user guide and API reference.
+
+The full Sphinx guide follows the Read the Docs style: [contents](https://github.com/SOULS-TAMU/Purin/blob/main/docs/index.rst),
+[runnable examples](https://github.com/SOULS-TAMU/Purin/blob/main/docs/examples/index.rst), [API reference](https://github.com/SOULS-TAMU/Purin/blob/main/docs/api/index.rst)
+and [build instructions](https://github.com/SOULS-TAMU/Purin/blob/main/docs/BUILD.md).
+
+| Resource | Contents |
+| --- | --- |
+| [Projection methods](#projection-methods) | Algorithm names, suitable problems and formulation choices |
+| [Contribution guide](https://github.com/SOULS-TAMU/Purin/blob/main/CONTRIBUTING.md) | Solver boundaries, numerical validation and packaging |
+| [Validation policy](https://github.com/SOULS-TAMU/Purin/blob/main/.github/validation/README.md) | API, gradient, sparse-kernel and quality checks |
+| [Changelog](https://github.com/SOULS-TAMU/Purin/blob/main/CHANGELOG.md) | Initial release scope and changes |
+
+## Contributing
+
+Bug reports, examples and focused pull requests are welcome. Read
+[CONTRIBUTING.md](https://github.com/SOULS-TAMU/Purin/blob/main/CONTRIBUTING.md) for setup, numerical checks and compatibility
+requirements. Keep generated results, environments and checkpoints out of Git.
+
+## Citation
+
+Use [CITATION.bib](https://github.com/SOULS-TAMU/Purin/blob/main/CITATION.bib) for a copyable BibTeX entry or GitHub's
+**Cite this repository** menu for [citation metadata](https://github.com/SOULS-TAMU/Purin/blob/main/CITATION.cff).
+
+## License
+
+[Apache License 2.0](https://github.com/SOULS-TAMU/Purin/blob/main/LICENSE). Authorship and upstream attribution are recorded
+in [AUTHORS](https://github.com/SOULS-TAMU/Purin/blob/main/AUTHORS) and [NOTICE](https://github.com/SOULS-TAMU/Purin/blob/main/NOTICE).
+
+[Changelog](https://github.com/SOULS-TAMU/Purin/blob/main/CHANGELOG.md) · [Report an issue](https://github.com/SOULS-TAMU/Purin/issues)