cog-suspect 2.1.1

Special Structure Detection for Pyomo

Special Structure Detection for Pyomo

This library implements methods to:

• Detect convex and concave expressions

• Detect increasing and decreasing expressions

• Detect linear, quadratic and polynomial expressions

• Tighten expression bounds

Please reference this software as

@Article{Suspect2019,
author={Ceccon, Francesco and Siirola, John D. and Misener, Ruth},
title={{SUSPECT}: {MINLP} special structure detector for Pyomo},
journal={Optimization Letters},
year={2019},
month={Feb},
issn="1862-4480",
doi="10.1007/s11590-019-01396-y",
url="https://doi.org/10.1007/s11590-019-01396-y"
}

Documentation

Documentation is available at https://cog-imperial.github.io/suspect/

Installation

SUSPECT requires Python 3.5 or later. We recommend installing SUSPECT in a virtual environment

To create the virtual environment run:

$ python3 -m venv myenv
$ source myenv/bin/activate

Then you are ready to clone and install SUSPECT:

$ git clone https://github.com/cog-imperial/suspect.git
$ cd suspect
$ pip install -r requirements.txt
$ pip install .

Command Line Usage

The package contains an utility to display structure information about a single problem.

You can run the utility as:

model_summary.py -p /path/to/problem.osil

or, if you want to check variables bounds include the solution:

model_summary.py -p /path/to/problem.osil -s /path/to/problem.sol

The repository also includes a Dockerfile to simplify running the utility in batch mode in a cloud environment. Refer to the batch folder for more information.

Library Usage

from suspect import detect_special_structure, create_connected_model
import pyomo.environ as aml


model = aml.ConcreteModel()
model.x = aml.Var()
model.y = aml.Var()

model.obj = aml.Objective(expr=(model.y - model.x)**3)
model.c1 = aml.Constraint(expr=model.y - model.x >= 0)

connected, _ = create_connected_model(model)
info = detect_special_structure(connected)

# try info.variables, info.objectives, and info.constraints
print(info.objectives['obj'])

License

Copyright 2020 Francesco Ceccon

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.

Acknowledgements

This work was funded by an Engineering & Physical Sciences Research Council Research Fellowship to RM [Grant Number EP/P016871/1].

Changelog

2.1.1 (2020-11-13)

• Try to compute special structure even if bounds are missing

2.1.0 (2020-09-16)

• Add support for log10

• Improve handling of quadratic expressions in nonlinear problems

• Fix FBBT bug when handling some types of expressions

2.0.2 (2020-09-01)

• Fix convexity on division

• Handle Pyomo MonomialTermExpression

2.0.1 (2020-07-01)

• Minor bug fixes

2.0.0 (2020-04-28)

• Use Pyomo expressions to represent the DAG

• Replace DAG with connected_model

1.6.0 (2019-11-15)

• Add floating point math mode

• Minor performance fixes

1.1.0 (2019-01-31)

• Add Quadratic expression type

• Add Interval special case for x*x

• Fix Interval sin

• Add Interval comparison with numbers

1.0.7 (2018-08-30)

• Add Interval abs

• Add Interval power

1.0.6 (2018-07-05)

• Change ExpressionType and UnaryFunctionType to IntEnum

1.0.5 (2018-07-05)

• Documentation improvements

1.0.4 (2018-07-04)

• First public release. Yay!