Installation

Prerequisites

  • Python 3.10.7

  • pip (modern version)

Installation

Standard Installation

Dyn-Benchmark is available from a dedicated package repository. You can install it using pip:

pip install dyn-benchmark[pretty]

For enhanced visualization capabilities, we recommend installing dyn-viz`:

# Install dyn-viz for visualization features
pip install dyn-viz

Installation from Source

For users who want to access the latest development version or contribute to the project, you can install directly from the source repository:

# Clone the repository (SSH)
git clone git@gitlab.com:decide.imt-atlantique/dyn/benchmark.git

# Or using HTTPS
git clone https://gitlab.com/decide.imt-atlantique/dyn/benchmark.git

# Install in development mode
pip install -e /path/to/dyn_benchmark

Using Virtual Environments

We recommend using virtual environments for installing Dyn-Benchmark to avoid conflicts with other packages.

Using venv:

python -m venv dyn-env
source dyn-env/bin/activate  # On Windows: dyn-env\Scripts\activate
# Then install the package as described above

Using pyenv with virtualenv:

pyenv install 3.10.7
pyenv virtualenv 3.10.7 dyn-env
pyenv local dyn-env
# Then install the package as described above

Using conda:

conda create -n dyn-env python=[3.10]
conda activate dyn-env
# Then install the package as described above

Verification

To verify your installation is working correctly, run the following Python code:

try:
    # Import the necessary components
    from dyn.benchmark.generator.groundtruth_generator import GroundtruthGenerator
    print("✓ Successfully imported the package")

    # Create a simple generator with minimal parameters
    generator = GroundtruthGenerator(seed=42)
    print("✓ Successfully created a generator")

    # Generate a very small benchmark (this will take a few seconds)
    print("Generating a small benchmark (this may take a few seconds)...")
    groundtruth = generator.generate()

    # If we reach this point, generation was successful
    print("✓ Successfully generated a benchmark with:")
    print(f"  - {len(groundtruth.graphs)} snapshots")
    print(f"  - {sum(len(g.nodes) for g in groundtruth.graphs.values())} total nodes across all snapshots")
    print(f"  - {sum(len(g.edges) for g in groundtruth.graphs.values())} total edges across all snapshots")
    print(f"  - {len(set(row.static_community_id for row in groundtruth.tcommlist))} static communities")
    print(f"  - {len(groundtruth.events)} community events")

    print("\nInstallation verified successfully!")

except Exception as e:
    print(f"\n❌ Verification failed: {type(e).__name__}: {e}")
    print("Please check the troubleshooting section or report this issue.")

Troubleshooting

If you encounter issues during installation:

  1. Ensure you’re using a compatible Python version

  2. Check that you have the necessary permissions to install packages

  3. If behind a proxy, configure pip appropriately

  4. For further assistance, please consult the troubleshoots document or create an issue on our GitLab repository