Installation

DynaSchedBench is distributed on PyPI as dsbx. The Python import name is dsbx, and every release command starts with dsbx-.

Supported Python Versions

The package metadata declares support for Python 3.9 through Python 3.12. Read the Docs builds this documentation with Python 3.11, as configured in .readthedocs.yaml.

Install From PyPI

python -m pip install dsbx

Check the installation:

python -c "import dsbx; print(dsbx.__version__)"
dsbx-gen --help

Install From Source

Use this path when contributing to the project, editing documentation, or running release checks locally.

git clone https://github.com/dsbx7/DynaSchedBench.git
cd DynaSchedBench
python -m pip install -e .[dev,docs]

The dev extra installs test and release tooling. The docs extra installs Sphinx, the Read the Docs theme, MyST Markdown support, and autodoc helpers.

Optional LLM Agent Configuration

The built-in heuristic and evolutionary agents do not require API keys. LLM agents read credentials from environment variables:

export DYNA_SCHEDBENCH_LLM_PROVIDER=openai
export DYNA_SCHEDBENCH_LLM_API_KEY=your_api_key
export DYNA_SCHEDBENCH_LLM_MODEL=gpt-4o-mini

You can also pass provider, model, base URL, sampling, and timeout options to dsbx-agent run. If no key is available, LLM agents fall back to a null client or heuristic behavior where implemented.

Verify Local Documentation Builds

From a source checkout:

python -m sphinx -W -b html docs docs/_build/html

Open docs/_build/html/index.html in a browser to inspect the generated site. The -W flag treats warnings as errors, matching the Read the Docs policy in this repository.