From 1aa87e61b7f43a3aa4e80aa13f1bbed160bb9795 Mon Sep 17 00:00:00 2001 From: Wouter Bles Date: Mon, 27 Apr 2026 16:33:03 +0200 Subject: [PATCH] docs: add conda instructions --- CHANGELOG.md | 1 + README.md | 18 +++++++++++++++++- 2 files changed, 18 insertions(+), 1 deletion(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index 81b4f9b..b4f0c4b 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -4,6 +4,7 @@ - Removed pre-v2 dead code (model.py, pyaugmecon.py, process_handler.py, queue_handler.py, solver_adapter.py, solver_process.py) superseded by the solver/ subpackage. + ## 2.0.0 - 2026-04-26 ### Breaking changes diff --git a/README.md b/README.md index 86a2ffe..68f6a7b 100644 --- a/README.md +++ b/README.md @@ -37,7 +37,7 @@ PyAUGMECON solves multi-objective optimization problems defined in [Pyomo](https ## Installation -PyAUGMECON does not bundle a solver. Pick one when installing: +PyAUGMECON is published on [PyPI](https://pypi.org/project/pyaugmecon) and [conda-forge](https://anaconda.org/conda-forge/pyaugmecon). The base package ships without a solver, so pick a solver extra when installing: ```bash # Recommended: uv (https://docs.astral.sh/uv/) @@ -45,6 +45,9 @@ uv add "pyaugmecon[highs]" # Or with pip pip install "pyaugmecon[highs]" + +# Or with conda / mamba +conda install -c conda-forge pyaugmecon ``` Available solver extras: @@ -65,6 +68,19 @@ Solvers not bundled as extras but supported when their binary is on `PATH`: Once installed, pass `solver_name="scip"` or `solver_name="cplex"`. You can also install CBC manually instead of `pyaugmecon[cbc]`; `cbcbox` is typically faster than a system-installed CBC binary (e.g. 85 s vs 471 s on `3kp40`), as it ships AVX2-optimized builds with bundled OpenBLAS. +> **conda solver names:** PyPI `[solver]` extras don't apply to conda. Install solvers as separate packages: +> +> | Solver | conda package | +> | --- | --- | +> | HiGHS | `conda install -c conda-forge highspy` | +> | Gurobi | `conda install -c gurobi gurobi` | +> | XPRESS | `conda install -c fico-xpress xpress` | +> | CPLEX | `conda install -c ibmdecisionoptimization cplex` | +> | CBC | `conda install -c conda-forge coincbc` | +> | SCIP | `conda install -c conda-forge scip` | +> +> Gurobi, XPRESS, and CPLEX are distributed via vendor-specific channels (not conda-forge) and require a license to use. + ## Quick start Define a Pyomo model with objectives in an `ObjectiveList` named `obj_list`, then pass it to PyAUGMECON. Defaults are sensible, so a minimal run takes no config: