opthub_runner 0.1.1

Library for running OptHub problems locally.

日本語版 👈

opthub-runner-python

opthub-runner-python is a Python package that provides Evaluator for local execution.

• Evaluator: Feature to evaluate solutions locally using Docker Image.

This repository describes how to install opthub-runner-python and use Evaluator.

Getting Started

First of all, you need to set up the following environment settings in advance.

• Install Python 3.10 or newer.
• Set up pip for package management.
• Install and start Docker.*

*For Mac users, you can install and start Docker Desktop.

After completing the above environment setup, you can install opthub-runner-python from PyPI with the following command.

pip install opthub-runner-python

After installation, you can use the Evaluator class by importing it with Python code as follows.

from opthub_runner.evaluator import Evaluator

For specific usage of the Evaluator class, please refer to Usage Example, Options, and Execution Results.

Note that Docker must be started to run the code. It will not run properly unless Docker is started.

Usage Example

Here is an example how to evaluate a solution using the Sphere function with the optimal solution as [1, 1] below.

from opthub_runner.evaluator import Evaluator



evaluator = Evaluator("opthub/sphere:latest",
                      {"SPHERE_OPTIMA": [[1, 1]]}) # Initialize Evaluator

x = [0, 2] # Solution to evaluate
result = evaluator.run(x) # Evaluate

print(result) # {'objective': 2, 'feasible': None, 'constraint': None, 'info': None}

In the initialization of the Evaluator, the Docker Image opthub/sphere:latest of the Sphere function and the environment variable SPHERE_OPTIMA of the Docker process are set. The run method then starts the Docker process internally and evaluates the solution [0, 2].

Information on the solutions input to the run method and the Docker image to be used are described in each problem on OptHub.

Options

The options used to initialize the Evaluator class are listed in the table below (* is required).

Option	Type	Default Value	Description
docker_image*	str	-	Name of the Docker Image used for evaluation.
environment*	dict[str, object]	-	Environment variables.
rm	bool	True	Whether to remove the Docker Container after evaluation.
timeout	int	43200	Timeout for evaluation using Docker Image.

Execution Results

The run method of the Evaluator class returns the evaluation result. The evaluation result is represented by a dictionary object containing pairs of Key and the corresponding Value. The following table shows the Keys and the types of the Values corresponding to the Keys included in the evaluation result.

Key	Type of Value	Description
objective	object	Objective function value of the solution.
feasible	bool or None	Feasibility of the solution.
constraint	object or None	Information on solution constraints.
info	object or None	Information on evaluation.

Detailed descriptions of objective, constraint, feasible, and info are provided in each problem on OptHub.

Troubleshooting

If you receive the following error, Docker is most likely not running. Please start Docker and re-run it. if you are Mac users, you can start Docker by launching Docker Desktop.

docker.errors.DockerException: Error while fetching server API version: ('Connection aborted.', FileNotFoundError(2, 'No such file or directory'))

For Contributors

Follow these steps to set up the environment:

1. Clone this repository.
2. Set up Poetry.
3. Run poetry install.
4. Download the recommended VSCode Extensions.
5. Disable the following VS Code Extensions for this workspace to avoid conflicts with other packages:
   • ms-python.pylint
   • ms-python.black-formatter
   • ms-python.flake8
   • ms-python.isort

Contact

If you have any questions or concerns, please feel free to contact us (Email: [e-mail]).