Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions .github/workflows/release.yml
Original file line number Diff line number Diff line change
Expand Up @@ -13,14 +13,14 @@ concurrency:
cancel-in-progress: false

env:
VAMOS_RELEASE_VERSION: "1.4.0"
VAMOS_RELEASE_VERSION: "1.5.0"
VAMOS_RELEASE_COMMIT: ${{ github.event.pull_request.head.sha || github.sha }}

jobs:
build-distributions:
if: >-
github.repository == 'vamos-optimization/VAMOS' &&
(github.event_name == 'workflow_dispatch' || github.head_ref == 'release/final-1.4.0')
(github.event_name == 'workflow_dispatch' || github.head_ref == 'release/final-1.5.0')
name: Build frozen distributions
runs-on: ubuntu-latest
steps:
Expand Down
4 changes: 2 additions & 2 deletions .github/workflows/upload_pypi.yml
Original file line number Diff line number Diff line change
Expand Up @@ -3,14 +3,14 @@ name: Publish frozen VAMOS release
on:
push:
tags:
- "v1.4.0"
- "v1.5.0"

permissions:
contents: read
actions: read

env:
VAMOS_RELEASE_VERSION: "1.4.0"
VAMOS_RELEASE_VERSION: "1.5.0"

jobs:
recover-frozen-artifacts:
Expand Down
6 changes: 3 additions & 3 deletions AGENTS.md
Original file line number Diff line number Diff line change
Expand Up @@ -101,7 +101,7 @@ Use the narrowest tier that proves the change while iterating, then run every hi
- Targeted: `python -m pytest -q <nearest-test-files>`, `python tools/typecheck.py --scope strict` for typed production changes, and `python tools/check_agent_docs.py` for agent/docs changes.
- Quick: `python -m pytest -q tests/test_check_agent_docs.py tests/architecture/test_docs_and_workflows.py tests/docs`.
- Full: `python tools/health.py`, `python -m pytest -q`, and `mkdocs build --strict`.
- Release: install the pinned build/release tools, then run the canonical `python tools/release_check.py --version 1.4.0`; it includes the full tier, stable/release typing, informational full-zero typing, frozen distribution inspection, installed-wheel smoke, dependency audits, SBOM, checksums, and provenance described in [Release verification](docs/release_smoke.md).
- Release: install the pinned build/release tools, then run the canonical `python tools/release_check.py --version 1.5.0`; it includes the full tier, stable/release typing, informational full-zero typing, frozen distribution inspection, installed-wheel smoke, dependency audits, SBOM, checksums, and provenance described in [Release verification](docs/release_smoke.md).

`tools/health.py` is the canonical local fast-fail architecture/tooling suite. CI has a distinct matrix and coverage scope. Both run exactly `python tools/check_agent_docs.py` for agent-documentation integrity; do not describe their complete suites as identical.

Expand Down Expand Up @@ -242,6 +242,6 @@ command: python tools/typecheck.py --scope full-zero
command: python -m pytest -q
command: mkdocs build --strict
command: python -m build
command: python tools/release_check.py --version 1.4.0
command: python tools/release_smoke.py --version 1.4.0 --mode full
command: python tools/release_check.py --version 1.5.0
command: python tools/release_smoke.py --version 1.5.0 --mode full
```
131 changes: 130 additions & 1 deletion CHANGELOG.md
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,134 @@ All notable public changes to VAMOS are documented here.

## [Unreleased]

## [1.5.0] - 2026-11-19

VAMOS 1.5.0 is a minor release. It adds AGE-MOEA-II as a variant of AGE-MOEA,
the authors' reading of RVEA's survival as a variant, one source of reference
vectors for MOEA/D, NSGA-III and RVEA (files, directories and two-layer
lattices), the CTP1–CTP7, IDTLZ1 and IDTLZ2 problems, DTLZ reference fronts with
4, 6 and 8 objectives, and the approximate hypervolume in the experimental
`vamos.ux.api`. It corrects NSGA-III's parent selection, which changes the
results of every NSGA-III run, and the packaged MOEA/D weight files, which
changes MOEA/D's results with four objectives and 100 to 119 individuals. The
unbounded external archive merges its rows when it is read, with the same
results. Seeded tuning runs of the NSGA-III and AGE-MOEA spaces draw different
trials. See the [1.5.0 release notes](docs/project/release-notes-1.5.0.md).

### Added

- Problems: CTP1–CTP7 (Deb, Pratap and Meyarivan, 2001), registered as
`ctp1`–`ctp7`, with the paper's definition and parameters.
- CTP1's J = 2 constraint coefficients come from the paper's procedure, and
CTP2–CTP7 from its constraint generator.
- g is ZDT4's Rastrigin function, with five variables by default.
- `vamos.problems.CTP(variant, n_var=..., n_constraints=..., theta=..., a=...,
b=..., c=..., d=..., e=...)` builds other instances.
- Two-objective reference fronts `ctp1.csv`–`ctp7.csv` are generated from the
definition.
- The constraint of CTP2–CTP7 is the paper's difference, not the NSGA-II
code's ratio that pymoo uses (#357).
- AGE-MOEA-II (Panichella, GECCO 2022) as a variant of AGE-MOEA:
- `AGEMOEAConfig.builder().variant("agemoea_ii")`, and `replacement:
agemoea_ii` in every AGE-MOEA tuning space (Evolver's `replacement:
agemoeaII`);
- its survival estimates the front's exponent by Newton-Raphson from the
point nearest to the ideal point, and measures diversity with geodesic
distances on the estimated manifold;
- its exponents, distances and survival scores are checked against jMetal's
`AGEMOEA2EnvironmentalSelection`;
- the default, `"agemoea"`, is unchanged; seeded tuning runs of the AGE-MOEA
spaces draw different trials (#359).
- Reference fronts: DTLZ1–DTLZ4 and DTLZ7 with 4, 6 and 8 objectives
(`dtlz1.<m>d.csv`, `dtlz2.<m>d.csv` and `dtlz7.<m>d.csv`), which
`resolve_reference_front_path` finds by `n_obj`; DTLZ3 and DTLZ4 resolve to
DTLZ2's file. DTLZ1–DTLZ4 use the Das-Dennis lattice of PlatEMO's reference
fronts, the points of jMetal's DTLZ2–DTLZ4 files with those numbers of
objectives (#362).
- Indicators: the experimental `vamos.ux.api` exports `HVIndicator`,
`HVApproxIndicator`, `HV_APPROX_METHODS` and `IndicatorResult`, documented in
Analysis and Visualization with a note on the exact hypervolume's cost from
about six objectives. `HVApproxIndicator.compute` passes `seed` and `method`
to `moocore.hv_approx` (an unknown method raises `ValueError`; `None` keeps
MooCore's default) and records `samples`, `seed` and the method used in
`details` (#363).
- MOEA/D, NSGA-III and RVEA take their vectors from one kind of option, as
jMetal and Evolver configure them: MOEA/D's `weight_vectors`, NSGA-III's
`reference_directions` and RVEA's new `reference_vectors` accept a file
(whitespace- or comma-separated) or a directory with jMetal's `W<m>D_<n>.dat`
names, the lattice of `divisions`, or, with the new `inner_divisions`, Deb and
Jain's two-layer lattice (156 vectors with eight objectives, 3 and 2
divisions). The vectors are the population, so RVEA can run 100 individuals
with three objectives from `W3D_100.dat`, and the three algorithms can start
from the same vectors. A path that does not exist raises
`ConfigurationError`. The resolved spec records a file for NSGA-III and RVEA
and the two-layer lattice for MOEA/D.
`vamos.algorithms.moead_populations(n_obj)` lists the populations MOEA/D runs
without weight options (#365).
- Problems: IDTLZ1 and IDTLZ2, the inverted DTLZ1 and DTLZ2 of Jain and Deb
(2014) as PlatEMO defines them, registered as `idtlz1` and `idtlz2` with a
configurable number of objectives (`n_var = n_obj + 4` and `n_obj + 9` by
default), with reference fronts for two and three objectives
(`idtlz1.2d.csv` to `idtlz2.3d.csv`) (#366).
- RVEA's `variant`: `"platemo"` runs the authors' reading of the survival, that
of PlatEMO's `RVEA/EnvironmentalSelection.m` and of jMetal (M equal to the
number of objectives in the angle penalty also with two objectives,
translation by each generation's minimum, vector adaptation to the
population's range); the default `"pymoo"` keeps VAMOS's results (#367).

### Changed

- An unbounded external archive keeps the rows each generation offers and
merges them (with moocore's non-dominated filter when moocore is installed)
only when the archive is read or the kept rows outnumber twice the archive,
instead of refiltering itself every generation.
- The rows, their order and every result are the same.
- A run that reads its archive only at the end filters it once: over 200 000
evaluations, 0.28 s instead of 85 s on a 112 000-row archive.
- `UnboundedArchive(buffered=False)` keeps the eager updates (#360).
- Two-layer lattices whose layers repeat a direction raise
`ConfigurationError`. A MOEA/D `weight_vectors` path that does not exist
raises `ConfigurationError` instead of being created as a directory for the
generated lattice, and an unknown key in `weight_vectors` or
`reference_directions` raises instead of being ignored (#365).

### Fixed

- MOEA/D with four objectives and 100 to 119 individuals runs the 100 weight
vectors of the packaged `W4D_100.dat`, which held them twice (as the jMetalPy
file it comes from) and was skipped, so VAMOS ran 84 individuals. The
packaged `W4D_165.dat` is space-separated, and weight files may separate
their components with whitespace or commas and depart from the simplex by
jMetal's rounding (`1e-5`), so all 31 packaged files load, also from the
command line, where 14 of them failed. NSGA-III reads
`reference_directions(path=...)` with the same reader, so jMetal's
whitespace-separated files such as `W3D_100.dat` work, and a path that does
not exist raises `ConfigurationError` instead of generating a lattice and
saving it there. Results change for MOEA/D with four objectives and 100 to
119 individuals; NSGA-III runs whose path was missing now fail (#361).
- NSGA-III selects its parents as configured, and never by crowding distance,
which is not part of NSGA-III. It ran a binary tournament on rank and
crowding distance, as NSGA-II, whatever its selection, so it was not the
published algorithm. `selection("random")`, now the default of
`NSGAIIIConfig.default()`, the command line and the tuning spaces, draws
parents at random, as Deb and Jain (2014); with constraints and feasibility
first, each parent wins a binary tournament on the constraint violation,
decided at random between feasible solutions, as Jain and Deb's constrained
NSGA-III and pymoo. `selection("tournament", size=k)` is jMetal's NSGA-III
tournament on rank, with ties drawn at random. Other selection methods and
options raise `ConfigurationError`. The NSGA-III tuning spaces choose
`selection`, `random` or `tournament` with `tournament_size`, as Evolver's,
and assignments of earlier spaces still build their tournament; seeded tuning
runs of these spaces draw different trials. `--selection-pressure` no longer
applies to NSGA-III. NSGA-III's results change, by default and with a
tournament, so experiments run with NSGA-III should be rerun (#364).
- A run whose MOEA/D `weight_vectors`, NSGA-III `reference_directions` or RVEA
`reference_vectors` names a file records the file it read and the `sha256` of
its vectors in its resolved configuration. Replay refuses the file if its
vectors have changed, where VAMOS 1.4.0 replayed whatever the file then held.
The run's record of the file is rebuilt from the stored configuration, so the
run verifies without the file (#365).

## [1.4.0] - 2026-11-17

VAMOS 1.4.0 is a minor release. The built-in tuning spaces become YAML grammars
Expand Down Expand Up @@ -1112,7 +1240,8 @@ development markers; they are not prior public releases.
- Studio, tuning, visualization, provider integrations, plugin interfaces, and
non-stable CLI commands remain experimental.

[Unreleased]: https://github.com/vamos-optimization/VAMOS/compare/v1.4.0...HEAD
[Unreleased]: https://github.com/vamos-optimization/VAMOS/compare/v1.5.0...HEAD
[1.5.0]: https://github.com/vamos-optimization/VAMOS/releases/tag/v1.5.0
[1.4.0]: https://github.com/vamos-optimization/VAMOS/releases/tag/v1.4.0
[1.3.0]: https://github.com/vamos-optimization/VAMOS/releases/tag/v1.3.0
[1.2.1]: https://github.com/vamos-optimization/VAMOS/releases/tag/v1.2.1
Expand Down
4 changes: 2 additions & 2 deletions CITATION.cff
Original file line number Diff line number Diff line change
Expand Up @@ -17,8 +17,8 @@ authors:
given-names: "J. Manuel"
affiliation: "Universidad Rey Juan Carlos, Spain"
title: "VAMOS: Vectorized Architecture for Multiobjective Optimization Studies"
version: 1.4.0
date-released: 2026-11-17
version: 1.5.0
date-released: 2026-11-19
url: "https://github.com/vamos-optimization/VAMOS"
preferred-citation:
type: article
Expand Down
9 changes: 7 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -24,7 +24,12 @@ results, rebuilds the tuning spaces and racing's statistics, and adds a
wall-clock time limit, problem options by name and jMetal's reference fronts as
an option. VAMOS 1.4.0 turns the tuning spaces into YAML grammars with defaults,
which the tuning backends and Studio read, gives MOEA/D and SMS-EMOA the choices
that Evolver offers, and removes VAMOS Assist and the LLM extras. See the
that Evolver offers, and removes VAMOS Assist and the LLM extras. VAMOS 1.5.0
corrects NSGA-III's parent selection, which changes its results, and the
packaged MOEA/D weight files, and adds AGE-MOEA-II, the authors' reading of
RVEA, one source of reference vectors for MOEA/D, NSGA-III and RVEA, the CTP and
IDTLZ problems and DTLZ reference fronts with up to eight objectives. See the
[1.5.0 release notes](https://vamos-optimization.org/project/release-notes-1.5.0/), the
[1.4.0 release notes](https://vamos-optimization.org/project/release-notes-1.4.0/), the
[1.3.0 release notes](https://vamos-optimization.org/project/release-notes-1.3.0/), the
[1.2.1 release notes](https://vamos-optimization.org/project/release-notes-1.2.1/), the
Expand Down Expand Up @@ -104,7 +109,7 @@ pip install "smac>=2.0"
| `numba` | Stable optional | Accelerates core numeric kernels such as variation, tournament selection, and MOEA/D neighborhood updates. |
| `moocore` | Stable optional | Adds accelerated quality-indicator support, especially hypervolume-style metrics. |

In VAMOS 1.4.0, use `engine="auto"` (CLI: `--engine auto`)
In VAMOS 1.5.0, use `engine="auto"` (CLI: `--engine auto`)
to prefer loadable Numba for all nine built-in algorithms, with a NumPy fallback.
Numba remains optional. Omitting `engine` still selects NumPy, preserving the 1.x
default; explicit `engine="numba"` fails if unavailable. First use can spend more
Expand Down
8 changes: 4 additions & 4 deletions docs/algorithms/agemoea.md
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@ downloads:

Adaptive geometry estimation uses an estimated non-Euclidean front geometry to support diversity selection in many-objective search [1]. Three-objective DTLZ2 provides a small, inspectable demonstration.

**Goal:** run the stable public API in VAMOS 1.4.0, interpret its returned population, and identify which settings control this mechanism.
**Goal:** run the stable public API in VAMOS 1.5.0, interpret its returned population, and identify which settings control this mechanism.

[Capability matrix](../reference/algorithms.md) · [Exact configuration reference](../reference/api/algorithms/agemoea.md) · [Setup and recorded evidence](../tutorials/coverage-setup.md)

Expand All @@ -31,7 +31,7 @@ The second command reruns the same protocol and requires Matplotlib. Use a new f
??? example "Complete runnable script"

```python
"""Verified VAMOS 1.4.0 lesson: agemoea. See coverage-setup for dependencies."""
"""Verified VAMOS 1.5.0 lesson: agemoea. See coverage-setup for dependencies."""


def execute_example(problem, algorithm, config, budget, seed=42, engine="numpy"):
Expand Down Expand Up @@ -157,7 +157,7 @@ Evaluations: 400

![Actual agemoea results from the documented protocol.](../assets/coverage/agemoea.svg)

*Measured with VAMOS 1.4.0; the [observations and provenance](../tutorials/coverage-setup.md#recorded-evidence) retain the inputs and results.*
*Measured with VAMOS 1.5.0; the [observations and provenance](../tutorials/coverage-setup.md#recorded-evidence) retain the inputs and results.*

## Understand the configuration

Expand All @@ -168,7 +168,7 @@ Evaluations: 400
| Polynomial mutation | Numeric probability `1/12`, eta 20. |
| `constraint_mode("feasibility")` | Enables feasibility-based selection if constraints exist; DTLZ2 here is unconstrained. |
| `constraint_handling(name, ...)` | **VAMOS 1.2.0:** ranks a constrained population with another of jMetal's techniques; its fronts take the place of the non-dominated ones in the survival and the tournament. See [choose a constraint-handling technique](../reference/constraints.md#choose-a-constraint-handling-technique). |
| `variant("agemoea_ii")` | **Development after 1.4.0:** runs AGE-MOEA-II [2], whose survival estimates the front's geometry by Newton-Raphson and measures diversity with geodesic distances; the default `"agemoea"` is the algorithm of this page. See [AGE-MOEA-II](../reference/algorithms.md#age-moea-ii). |
| `variant("agemoea_ii")` | **VAMOS 1.5.0:** runs AGE-MOEA-II [2], whose survival estimates the front's geometry by Newton-Raphson and measures diversity with geodesic distances; the default `"agemoea"` is the algorithm of this page. See [AGE-MOEA-II](../reference/algorithms.md#age-moea-ii). |

## Interpret the result and its limits

Expand Down
6 changes: 3 additions & 3 deletions docs/algorithms/ibea.md
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@ downloads:

Indicator-based selection turns pairwise objective comparisons into fitness [1]. Here the additive epsilon indicator drives search; the indicator chosen for selection need not be the metric used to report final quality.

**Goal:** run the stable public API in VAMOS 1.4.0, interpret its returned population, and identify which settings control this mechanism.
**Goal:** run the stable public API in VAMOS 1.5.0, interpret its returned population, and identify which settings control this mechanism.

[Capability matrix](../reference/algorithms.md) · [Exact configuration reference](../reference/api/algorithms/ibea.md) · [Setup and recorded evidence](../tutorials/coverage-setup.md)

Expand All @@ -31,7 +31,7 @@ The second command reruns the same protocol and requires Matplotlib. Use a new f
??? example "Complete runnable script"

```python
"""Verified VAMOS 1.4.0 lesson: ibea. See coverage-setup for dependencies."""
"""Verified VAMOS 1.5.0 lesson: ibea. See coverage-setup for dependencies."""


def execute_example(problem, algorithm, config, budget, seed=42, engine="numpy"):
Expand Down Expand Up @@ -159,7 +159,7 @@ Evaluations: 400

![Actual ibea results from the documented protocol.](../assets/coverage/ibea.svg)

*Measured with VAMOS 1.4.0; the [observations and provenance](../tutorials/coverage-setup.md#recorded-evidence) retain the inputs and results.*
*Measured with VAMOS 1.5.0; the [observations and provenance](../tutorials/coverage-setup.md#recorded-evidence) retain the inputs and results.*

## Understand the configuration

Expand Down
2 changes: 1 addition & 1 deletion docs/algorithms/moead.md
Original file line number Diff line number Diff line change
Expand Up @@ -84,7 +84,7 @@ For two objectives, `MOEADConfig.default(pop_size=100, n_obj=2, ...)` requests 9
- **Packaged weights:** with three or more objectives and a `pop_size` that no complete lattice matches, `MOEADConfig.default(pop_size=...)` and `optimize` use the weight file VAMOS packages for that size, as jMetal does (for example `W3D_100.dat`, so 100 individuals with three objectives work). In VAMOS 1.1.0 this failed with a lattice-size error.
- **Defaults:** the default DE crossover with PBI aggregation is a VAMOS choice; `aggregation("tchebycheff")` with the other defaults reproduces jMetal's MOEA/D-DE.

**Development after 1.4.0:** four objectives and 100 individuals run the packaged `W4D_100.dat`, which held its vectors twice and was skipped, so VAMOS 1.4.0 ran 84 individuals for any population from 100 to 119. Every packaged weight file now loads, also from the command line, and weight files may separate their components by whitespace or commas; see the [configuration reference](../reference/api/algorithms/moead.md).
**VAMOS 1.5.0:** four objectives and 100 individuals run the packaged `W4D_100.dat`, which held its vectors twice and was skipped, so VAMOS 1.4.0 ran 84 individuals for any population from 100 to 119. Every packaged weight file now loads, also from the command line, and weight files may separate their components by whitespace or commas; see the [configuration reference](../reference/api/algorithms/moead.md).

**VAMOS 1.4.0:** MOEA/D gains the choices of Evolver's MOEA/D, with jMetal's definitions. Each is off by default, and a configuration without them runs as in 1.3.0:

Expand Down
Loading
Loading