mmahelpy 0.1.0__tar.gz

mmahelpy-0.1.0/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 [yyyy] [name of copyright owner]
+
+   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.

mmahelpy-0.1.0/PKG-INFO

@@ -0,0 +1,90 @@
+Metadata-Version: 2.4
+Name: mmahelpy
+Version: 0.1.0
+Summary: A reusable Python implementation of the Method of Moving Asymptotes
+Author: Kevin Tofu
+License-Expression: Apache-2.0
+Project-URL: Repository, https://github.com/kevin-tofu/mmahelpy
+Project-URL: Issues, https://github.com/kevin-tofu/mmahelpy/issues
+Keywords: mma,moving-asymptotes,optimization
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Science/Research
+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: Topic :: Scientific/Engineering :: Mathematics
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy>=1.24
+Requires-Dist: scipy>=1.10
+Provides-Extra: test
+Requires-Dist: pytest>=8; extra == "test"
+Dynamic: license-file
+
+# mmapy source package
+
+`mmapy` is a small, stateful implementation of the Method of Moving
+Asymptotes (MMA). It solves a separable convex approximation at each call and
+retains the design and asymptote history for the next outer iteration.
+
+The package is independent of finite-element and topology-optimization
+libraries. Supply objective and inequality-constraint values and gradients at
+the current point. Constraints use the convention `g(x) <= 0`.
+
+## Install
+
+```bash
+pip install mmahelpy
+```
+
+## Example
+
+```python
+import numpy as np
+from mmapy import MMA, MMAOptions
+
+optimizer = MMA(MMAOptions(move_limit=0.2))
+x = np.array([0.8, 0.8])
+lower = np.zeros(2)
+upper = np.ones(2)
+
+for _ in range(30):
+    # Minimize x[0]**2 + x[1]**2 subject to x[0] + x[1] >= 0.9.
+    objective = float(x @ x)
+    objective_gradient = 2.0 * x
+    constraints = np.array([0.9 - x.sum()])
+    constraint_jacobian = np.array([[-1.0, -1.0]])
+    result = optimizer.step(
+        x,
+        objective,
+        objective_gradient,
+        constraints,
+        constraint_jacobian,
+        lower,
+        upper,
+    )
+    x = result.x
+
+print(x)
+```
+
+Create one `MMA` object per optimization run. Its `step` method returns an
+`MMAResult` with the next design, dual multipliers, approximate constraint
+values, and dual-solver status. Call `reset()` to reuse the object for a new
+problem. Multiple inequality constraints are supported.
+
+## Development
+
+```bash
+PYTHONPATH=src pytest
+```
+
+The distribution is named `mmahelpy` because the `mmapy` name is already
+used on PyPI. The installed import remains `mmapy`. Scikit-Topt depends on
+`mmahelpy` and adapts its public `MMA_Config` to the generic optimizer for
+topology optimization. The solver focuses on continuous, differentiable
+problems with box bounds and inequality constraints.

mmahelpy-0.1.0/README.md

@@ -0,0 +1,63 @@
+# mmapy source package
+
+`mmapy` is a small, stateful implementation of the Method of Moving
+Asymptotes (MMA). It solves a separable convex approximation at each call and
+retains the design and asymptote history for the next outer iteration.
+
+The package is independent of finite-element and topology-optimization
+libraries. Supply objective and inequality-constraint values and gradients at
+the current point. Constraints use the convention `g(x) <= 0`.
+
+## Install
+
+```bash
+pip install mmahelpy
+```
+
+## Example
+
+```python
+import numpy as np
+from mmapy import MMA, MMAOptions
+
+optimizer = MMA(MMAOptions(move_limit=0.2))
+x = np.array([0.8, 0.8])
+lower = np.zeros(2)
+upper = np.ones(2)
+
+for _ in range(30):
+    # Minimize x[0]**2 + x[1]**2 subject to x[0] + x[1] >= 0.9.
+    objective = float(x @ x)
+    objective_gradient = 2.0 * x
+    constraints = np.array([0.9 - x.sum()])
+    constraint_jacobian = np.array([[-1.0, -1.0]])
+    result = optimizer.step(
+        x,
+        objective,
+        objective_gradient,
+        constraints,
+        constraint_jacobian,
+        lower,
+        upper,
+    )
+    x = result.x
+
+print(x)
+```
+
+Create one `MMA` object per optimization run. Its `step` method returns an
+`MMAResult` with the next design, dual multipliers, approximate constraint
+values, and dual-solver status. Call `reset()` to reuse the object for a new
+problem. Multiple inequality constraints are supported.
+
+## Development
+
+```bash
+PYTHONPATH=src pytest
+```
+
+The distribution is named `mmahelpy` because the `mmapy` name is already
+used on PyPI. The installed import remains `mmapy`. Scikit-Topt depends on
+`mmahelpy` and adapts its public `MMA_Config` to the generic optimizer for
+topology optimization. The solver focuses on continuous, differentiable
+problems with box bounds and inequality constraints.

mmahelpy-0.1.0/pyproject.toml

@@ -0,0 +1,42 @@
+[build-system]
+requires = ["setuptools>=61", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "mmahelpy"
+version = "0.1.0"
+description = "A reusable Python implementation of the Method of Moving Asymptotes"
+readme = "README.md"
+requires-python = ">=3.10"
+license = "Apache-2.0"
+authors = [{ name = "Kevin Tofu" }]
+dependencies = ["numpy>=1.24", "scipy>=1.10"]
+keywords = ["mma", "moving-asymptotes", "optimization"]
+classifiers = [
+  "Development Status :: 3 - Alpha",
+  "Intended Audience :: Science/Research",
+  "Programming Language :: Python :: 3",
+  "Programming Language :: Python :: 3 :: Only",
+  "Programming Language :: Python :: 3.10",
+  "Programming Language :: Python :: 3.11",
+  "Programming Language :: Python :: 3.12",
+  "Programming Language :: Python :: 3.13",
+  "Topic :: Scientific/Engineering :: Mathematics",
+]
+
+[project.urls]
+Repository = "https://github.com/kevin-tofu/mmahelpy"
+Issues = "https://github.com/kevin-tofu/mmahelpy/issues"
+
+[project.optional-dependencies]
+test = ["pytest>=8"]
+
+[tool.setuptools]
+package-dir = {"" = "src"}
+
+[tool.setuptools.packages.find]
+where = ["src"]
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]
+addopts = "-q"

mmahelpy-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

mmahelpy-0.1.0/src/mmahelpy.egg-info/PKG-INFO

@@ -0,0 +1,90 @@
+Metadata-Version: 2.4
+Name: mmahelpy
+Version: 0.1.0
+Summary: A reusable Python implementation of the Method of Moving Asymptotes
+Author: Kevin Tofu
+License-Expression: Apache-2.0
+Project-URL: Repository, https://github.com/kevin-tofu/mmahelpy
+Project-URL: Issues, https://github.com/kevin-tofu/mmahelpy/issues
+Keywords: mma,moving-asymptotes,optimization
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Science/Research
+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: Topic :: Scientific/Engineering :: Mathematics
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy>=1.24
+Requires-Dist: scipy>=1.10
+Provides-Extra: test
+Requires-Dist: pytest>=8; extra == "test"
+Dynamic: license-file
+
+# mmapy source package
+
+`mmapy` is a small, stateful implementation of the Method of Moving
+Asymptotes (MMA). It solves a separable convex approximation at each call and
+retains the design and asymptote history for the next outer iteration.
+
+The package is independent of finite-element and topology-optimization
+libraries. Supply objective and inequality-constraint values and gradients at
+the current point. Constraints use the convention `g(x) <= 0`.
+
+## Install
+
+```bash
+pip install mmahelpy
+```
+
+## Example
+
+```python
+import numpy as np
+from mmapy import MMA, MMAOptions
+
+optimizer = MMA(MMAOptions(move_limit=0.2))
+x = np.array([0.8, 0.8])
+lower = np.zeros(2)
+upper = np.ones(2)
+
+for _ in range(30):
+    # Minimize x[0]**2 + x[1]**2 subject to x[0] + x[1] >= 0.9.
+    objective = float(x @ x)
+    objective_gradient = 2.0 * x
+    constraints = np.array([0.9 - x.sum()])
+    constraint_jacobian = np.array([[-1.0, -1.0]])
+    result = optimizer.step(
+        x,
+        objective,
+        objective_gradient,
+        constraints,
+        constraint_jacobian,
+        lower,
+        upper,
+    )
+    x = result.x
+
+print(x)
+```
+
+Create one `MMA` object per optimization run. Its `step` method returns an
+`MMAResult` with the next design, dual multipliers, approximate constraint
+values, and dual-solver status. Call `reset()` to reuse the object for a new
+problem. Multiple inequality constraints are supported.
+
+## Development
+
+```bash
+PYTHONPATH=src pytest
+```
+
+The distribution is named `mmahelpy` because the `mmapy` name is already
+used on PyPI. The installed import remains `mmapy`. Scikit-Topt depends on
+`mmahelpy` and adapts its public `MMA_Config` to the generic optimizer for
+topology optimization. The solver focuses on continuous, differentiable
+problems with box bounds and inequality constraints.

mmahelpy-0.1.0/src/mmahelpy.egg-info/SOURCES.txt

@@ -0,0 +1,11 @@
+LICENSE
+README.md
+pyproject.toml
+src/mmahelpy.egg-info/PKG-INFO
+src/mmahelpy.egg-info/SOURCES.txt
+src/mmahelpy.egg-info/dependency_links.txt
+src/mmahelpy.egg-info/requires.txt
+src/mmahelpy.egg-info/top_level.txt
+src/mmapy/__init__.py
+src/mmapy/mma.py
+tests/test_mma.py

mmahelpy-0.1.0/src/mmahelpy.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

mmahelpy-0.1.0/src/mmahelpy.egg-info/requires.txt

@@ -0,0 +1,5 @@
+numpy>=1.24
+scipy>=1.10
+
+[test]
+pytest>=8

mmahelpy-0.1.0/src/mmahelpy.egg-info/top_level.txt

@@ -0,0 +1 @@
+mmapy

mmahelpy-0.1.0/src/mmapy/__init__.py

@@ -0,0 +1,5 @@
+"""Stateful Method of Moving Asymptotes optimizer used by Scikit-Topt."""
+
+from mmapy.mma import MMA, MMAOptions, MMAResult
+
+__all__ = ["MMA", "MMAOptions", "MMAResult"]

mmahelpy-0.1.0/src/mmapy/mma.py

@@ -0,0 +1,315 @@
+"""A stateful, separable-convex MMA optimizer with multiple constraints."""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+
+import numpy as np
+from numpy.typing import ArrayLike, NDArray
+from scipy.optimize import minimize
+
+
+FloatArray = NDArray[np.float64]
+
+
+@dataclass(frozen=True)
+class MMAOptions:
+    """Numerical settings for the MMA outer iteration."""
+
+    move_limit: float | None = None
+    initial_asymptote: float = 0.5
+    asymptote_increase: float = 1.2
+    asymptote_decrease: float = 0.7
+    asymptote_minimum: float = 0.01
+    asymptote_maximum: float = 10.0
+    regularization: float = 1e-5
+    dual_tolerance: float = 1e-9
+    dual_max_iterations: int = 200
+
+    def __post_init__(self) -> None:
+        if self.move_limit is not None and self.move_limit <= 0.0:
+            raise ValueError("move_limit must be positive")
+        if self.initial_asymptote <= 0.0:
+            raise ValueError("initial_asymptote must be positive")
+        if self.asymptote_increase < 1.0:
+            raise ValueError("asymptote_increase must be at least one")
+        if not 0.0 < self.asymptote_decrease <= 1.0:
+            raise ValueError("asymptote_decrease must be in (0, 1]")
+        if not 0.0 < self.asymptote_minimum < self.asymptote_maximum:
+            raise ValueError("asymptote limits must satisfy 0 < min < max")
+        if self.regularization <= 0.0:
+            raise ValueError("regularization must be positive")
+        if self.dual_tolerance <= 0.0:
+            raise ValueError("dual_tolerance must be positive")
+        if self.dual_max_iterations < 1:
+            raise ValueError("dual_max_iterations must be positive")
+
+
+@dataclass(frozen=True)
+class MMAResult:
+    """Result of one MMA subproblem solve."""
+
+    x: FloatArray
+    multipliers: FloatArray
+    approximated_constraints: FloatArray
+    dual_iterations: int
+    dual_success: bool
+
+
+def _asymptote_coefficients(
+    gradient: FloatArray,
+    x: FloatArray,
+    lower: FloatArray,
+    upper: FloatArray,
+    regularization: float,
+) -> tuple[FloatArray, FloatArray]:
+    """Return reciprocal coefficients matching value and gradient at x."""
+    span = upper - lower
+    p = np.maximum(gradient, 0.0)
+    q = np.maximum(-gradient, 0.0)
+    stabilizer = 0.001 * (p + q) + regularization / span
+    return (
+        (p + stabilizer) * (upper - x) ** 2,
+        (q + stabilizer) * (x - lower) ** 2,
+    )
+
+
+def _approximation(
+    x: FloatArray,
+    x0: FloatArray,
+    value0: FloatArray,
+    p: FloatArray,
+    q: FloatArray,
+    lower: FloatArray,
+    upper: FloatArray,
+) -> FloatArray:
+    """Evaluate one or more reciprocal MMA approximations."""
+    return value0 + np.sum(
+        p * (1.0 / (upper - x) - 1.0 / (upper - x0))
+        + q * (1.0 / (x - lower) - 1.0 / (x0 - lower)),
+        axis=-1,
+    )
+
+
+class MMA:
+    """Stateful MMA optimizer for differentiable bound-constrained problems.
+
+    Call :meth:`step` once per outer iteration with the objective and
+    inequality constraints evaluated at the current design. Constraint values
+    must use the convention ``g(x) <= 0``. The optimizer retains design and
+    asymptote history between calls.
+
+    Parameters
+    ----------
+    config:
+        Moving-asymptote and dual-solver settings.
+    """
+
+    def __init__(self, config: MMAOptions | None = None) -> None:
+        self.config = config or MMAOptions()
+        self._previous_x: FloatArray | None = None
+        self._previous_previous_x: FloatArray | None = None
+        self._lower_asymptote: FloatArray | None = None
+        self._upper_asymptote: FloatArray | None = None
+
+    def reset(self) -> None:
+        """Clear iteration history before starting a new optimization."""
+        self._previous_x = None
+        self._previous_previous_x = None
+        self._lower_asymptote = None
+        self._upper_asymptote = None
+
+    def step(
+        self,
+        x: ArrayLike,
+        objective: float,
+        objective_gradient: ArrayLike,
+        constraints: ArrayLike,
+        constraint_jacobian: ArrayLike,
+        lower_bounds: ArrayLike,
+        upper_bounds: ArrayLike,
+        *,
+        move_limit: float | ArrayLike | None = None,
+    ) -> MMAResult:
+        """Solve one local MMA approximation and update optimizer history.
+
+        ``constraint_jacobian`` has shape ``(n_constraints, n_variables)``.
+        Empty one-dimensional inputs are accepted when there are no
+        constraints. Bounds are global design bounds; an optional move limit
+        further restricts this iteration's step.
+        """
+        x0 = np.asarray(x, dtype=float)
+        df0 = np.asarray(objective_gradient, dtype=float)
+        g0 = np.atleast_1d(np.asarray(constraints, dtype=float))
+        dg0 = np.asarray(constraint_jacobian, dtype=float)
+        xmin = np.broadcast_to(np.asarray(lower_bounds, dtype=float), x0.shape)
+        xmax = np.broadcast_to(np.asarray(upper_bounds, dtype=float), x0.shape)
+        if x0.ndim != 1 or x0.size == 0:
+            raise ValueError("x must be a nonempty one-dimensional array")
+        n = x0.size
+        if df0.shape != (n,):
+            raise ValueError("objective_gradient must have shape (n_variables,)")
+        if xmin.shape != (n,) or xmax.shape != (n,):
+            raise ValueError("bounds must broadcast to x")
+        if (
+            np.any(~np.isfinite(x0))
+            or np.any(~np.isfinite(df0))
+            or np.any(~np.isfinite(xmin))
+            or np.any(~np.isfinite(xmax))
+        ):
+            raise ValueError("x, gradients, and bounds must be finite")
+        if np.any(~np.isfinite(g0)) or np.any(~np.isfinite(dg0)):
+            raise ValueError("constraints and constraint_jacobian must be finite")
+        if np.any(xmin >= xmax) or np.any(x0 < xmin) or np.any(x0 > xmax):
+            raise ValueError("x must lie within nondegenerate global bounds")
+
+        m = g0.size
+        if m == 0:
+            dg0 = np.empty((0, n), dtype=float)
+        elif dg0.shape != (m, n):
+            raise ValueError(
+                "constraint_jacobian must have shape (n_constraints, n_variables)"
+            )
+        if not np.isfinite(objective):
+            raise ValueError("objective must be finite")
+
+        cfg = self.config
+        span = xmax - xmin
+        low, upp = self._moving_asymptotes(x0, span)
+        move = cfg.move_limit if move_limit is None else move_limit
+        if move is None:
+            move_array = span
+        else:
+            move_array = np.broadcast_to(np.asarray(move, dtype=float), x0.shape)
+            if np.any(~np.isfinite(move_array)) or np.any(move_array <= 0.0):
+                raise ValueError("move_limit must contain finite positive values")
+        alpha = np.maximum.reduce((
+            xmin,
+            low + 0.1 * (x0 - low),
+            x0 - move_array,
+        ))
+        beta = np.minimum.reduce((
+            xmax,
+            upp - 0.1 * (upp - x0),
+            x0 + move_array,
+        ))
+        if np.any(alpha >= beta):
+            raise ValueError("MMA move limits leave no feasible step")
+
+        p0, q0 = _asymptote_coefficients(
+            df0, x0, low, upp, cfg.regularization
+        )
+        if m:
+            p_constraints = np.empty((m, n), dtype=float)
+            q_constraints = np.empty((m, n), dtype=float)
+            for i in range(m):
+                p_constraints[i], q_constraints[i] = _asymptote_coefficients(
+                    dg0[i], x0, low, upp, cfg.regularization
+                )
+        else:
+            p_constraints = np.empty((0, n), dtype=float)
+            q_constraints = np.empty((0, n), dtype=float)
+
+        def primal(multipliers: FloatArray) -> FloatArray:
+            p = p0 + multipliers @ p_constraints if m else p0
+            q = q0 + multipliers @ q_constraints if m else q0
+            sqrt_p = np.sqrt(np.maximum(p, np.finfo(float).tiny))
+            sqrt_q = np.sqrt(np.maximum(q, np.finfo(float).tiny))
+            unconstrained = (
+                sqrt_q * upp + sqrt_p * low
+            ) / (sqrt_p + sqrt_q)
+            return np.clip(unconstrained, alpha, beta)
+
+        def objective_approximation(y: FloatArray) -> float:
+            return float(_approximation(
+                y, x0, np.asarray(objective), p0, q0, low, upp
+            ))
+
+        def constraints_approximation(y: FloatArray) -> FloatArray:
+            if not m:
+                return np.empty(0, dtype=float)
+            return _approximation(
+                y, x0, g0, p_constraints, q_constraints, low, upp
+            )
+
+        if m:
+            def negative_dual(lam: FloatArray) -> tuple[float, FloatArray]:
+                y = primal(lam)
+                g_approx = constraints_approximation(y)
+                value = objective_approximation(y) + float(lam @ g_approx)
+                return -value, -g_approx
+
+            dual = minimize(
+                negative_dual,
+                np.zeros(m, dtype=float),
+                method="L-BFGS-B",
+                jac=True,
+                bounds=[(0.0, None)] * m,
+                options={
+                    "ftol": cfg.dual_tolerance,
+                    "gtol": cfg.dual_tolerance,
+                    "maxiter": cfg.dual_max_iterations,
+                    "maxls": 40,
+                },
+            )
+            multipliers = np.asarray(dual.x, dtype=float)
+            x_new = primal(multipliers)
+            approximated_constraints = constraints_approximation(x_new)
+            dual_iterations = int(dual.nit)
+            dual_success = bool(dual.success)
+        else:
+            multipliers = np.empty(0, dtype=float)
+            x_new = primal(multipliers)
+            approximated_constraints = np.empty(0, dtype=float)
+            dual_iterations = 0
+            dual_success = True
+
+        self._advance_history(x0, low, upp)
+        return MMAResult(
+            x=x_new.copy(),
+            multipliers=multipliers.copy(),
+            approximated_constraints=approximated_constraints.copy(),
+            dual_iterations=dual_iterations,
+            dual_success=dual_success,
+        )
+
+    def _moving_asymptotes(
+        self,
+        x: FloatArray,
+        span: FloatArray,
+    ) -> tuple[FloatArray, FloatArray]:
+        cfg = self.config
+        if self._lower_asymptote is None or self._upper_asymptote is None:
+            distance = cfg.initial_asymptote * span
+            return x - distance, x + distance
+
+        factor = np.ones_like(x)
+        if self._previous_x is not None and self._previous_previous_x is not None:
+            trend = (x - self._previous_x) * (
+                self._previous_x - self._previous_previous_x
+            )
+            factor[trend > 0.0] = cfg.asymptote_increase
+            factor[trend < 0.0] = cfg.asymptote_decrease
+
+        lower = x - factor * (self._previous_x - self._lower_asymptote)
+        upper = x + factor * (self._upper_asymptote - self._previous_x)
+        min_distance = cfg.asymptote_minimum * span
+        max_distance = cfg.asymptote_maximum * span
+        lower = np.minimum(x - min_distance, np.maximum(x - max_distance, lower))
+        upper = np.maximum(x + min_distance, np.minimum(x + max_distance, upper))
+        return lower, upper
+
+    def _advance_history(
+        self,
+        x: FloatArray,
+        lower: FloatArray,
+        upper: FloatArray,
+    ) -> None:
+        if self._previous_x is not None:
+            self._previous_previous_x = self._previous_x.copy()
+        self._previous_x = x.copy()
+        self._lower_asymptote = lower.copy()
+        self._upper_asymptote = upper.copy()
+
+
+__all__ = ["MMA", "MMAOptions", "MMAResult"]

mmahelpy-0.1.0/tests/test_mma.py

@@ -0,0 +1,101 @@
+import numpy as np
+
+from mmapy import MMA, MMAOptions
+
+
+def test_mma_solves_a_linear_single_constraint_subproblem():
+    optimizer = MMA(MMAOptions(move_limit=0.4))
+    result = optimizer.step(
+        x=[0.8],
+        objective=0.8,
+        objective_gradient=[1.0],
+        constraints=[0.5 - 0.8],
+        constraint_jacobian=[[-1.0]],
+        lower_bounds=[0.0],
+        upper_bounds=[1.0],
+    )
+
+    assert 0.5 <= result.x[0] < 0.8
+    assert abs(result.approximated_constraints[0]) < 1e-6
+    assert result.multipliers[0] >= 0.0
+    assert result.dual_success
+
+
+def test_mma_supports_multiple_inequality_constraints():
+    optimizer = MMA(MMAOptions(move_limit=0.4))
+    result = optimizer.step(
+        x=[0.8, 0.8],
+        objective=1.6,
+        objective_gradient=[1.0, 1.0],
+        constraints=[0.4 - 0.8, 0.6 - 0.8],
+        constraint_jacobian=[[-1.0, 0.0], [0.0, -1.0]],
+        lower_bounds=[0.0, 0.0],
+        upper_bounds=[1.0, 1.0],
+    )
+
+    assert result.x[0] >= 0.4
+    assert result.x[1] >= 0.6
+    assert np.all(np.abs(result.approximated_constraints) < 2e-3)
+    assert result.multipliers.shape == (2,)
+
+
+def test_mma_converges_on_a_quadratic_problem_with_a_linear_constraint():
+    optimizer = MMA(MMAOptions(move_limit=0.2))
+    x = np.array([0.8, 0.8])
+    for _ in range(30):
+        result = optimizer.step(
+            x=x,
+            objective=float(x @ x),
+            objective_gradient=2.0 * x,
+            constraints=[0.9 - x.sum()],
+            constraint_jacobian=[[-1.0, -1.0]],
+            lower_bounds=[0.0, 0.0],
+            upper_bounds=[1.0, 1.0],
+        )
+        assert result.dual_success
+        x = result.x
+
+    np.testing.assert_allclose(x, [0.45, 0.45], atol=1e-5)
+    assert x.sum() >= 0.9 - 1e-6
+
+
+def test_mma_keeps_asymptote_history_and_reset_clears_it():
+    optimizer = MMA(MMAOptions(move_limit=0.2))
+    args = dict(
+        objective=0.0,
+        objective_gradient=[-1.0],
+        constraints=[],
+        constraint_jacobian=np.empty((0, 1)),
+        lower_bounds=[0.0],
+        upper_bounds=[1.0],
+    )
+    first = optimizer.step(x=[0.5], **args)
+    assert optimizer._lower_asymptote is not None
+    assert optimizer._upper_asymptote is not None
+
+    optimizer.step(x=first.x, **args)
+    assert optimizer._previous_previous_x is not None
+
+    optimizer.reset()
+    assert optimizer._previous_x is None
+    assert optimizer._previous_previous_x is None
+    assert optimizer._lower_asymptote is None
+    assert optimizer._upper_asymptote is None
+
+
+def test_mma_rejects_inconsistent_constraint_jacobian_shape():
+    optimizer = MMA()
+    try:
+        optimizer.step(
+            x=[0.5, 0.5],
+            objective=1.0,
+            objective_gradient=[1.0, 1.0],
+            constraints=[-0.1],
+            constraint_jacobian=[[1.0]],
+            lower_bounds=[0.0, 0.0],
+            upper_bounds=[1.0, 1.0],
+        )
+    except ValueError as error:
+        assert "constraint_jacobian" in str(error)
+    else:
+        raise AssertionError("shape mismatch should have raised ValueError")