Installation Guide¶

This guide covers all the ways to install and set up the TSG Ecosystem components.

System Requirements¶

Python 3.10 or higher
4GB RAM minimum (8GB recommended for larger datasets)
Operating systems:
Windows 10 or higher
macOS 10.14 or higher
Linux (most modern distributions)

Standard Installation¶

The simplest way to install the TSG Ecosystem is through pip:

pip install tsgraph

This installs the core library with basic functionality.

Component-specific Installation¶

You can install specific components of the ecosystem:

# Core library only
pip install tsgeco-core

# Visualization components
pip install tsgeco-viz

# Analytics suite
pip install tsgeco-analytics

# Converter utilities
pip install tsgeco-converter

Optional Dependencies¶

Install optional features using extras:

# Install with visualization dependencies
pip install tsgeco[viz]

# Install with analytics dependencies
pip install tsgeco[analytics]

# Install with all optional dependencies
pip install tsgeco[all]

Development Installation¶

For contributors who want to install the development version:

git clone https://github.com/TSGECO/tsgeco.git
cd tsgeco
pip install -e ".[dev]"

This installs the package in development mode with additional tools for testing and development.

GPU Support¶

For enhanced performance with large datasets, we recommend installing with GPU support:

pip install tsgeco[gpu]

CUDA Requirements¶

If using the GPU-accelerated features: - CUDA 11.0 or higher - Compatible NVIDIA GPU with at least 4GB memory

Docker Installation¶

We provide Docker images with all dependencies pre-installed:

# Pull the latest image
docker pull tsgeco/tsgeco:latest

# Or a specific version
docker pull tsgeco/tsgeco:1.0.0

# Run a container
docker run -it --rm tsgeco/tsgeco

Verification¶

After installation, you can verify that everything is working correctly:

python -c "import tsgeco; print(tsgeco.__version__)"

You should see the version number of the installed package.

Troubleshooting¶

Common Issues¶

Missing Dependencies¶

If you encounter errors about missing dependencies, try:

pip install --upgrade tsgeco[all]

GPU Not Detected¶

If your GPU isn't being detected:

Verify your CUDA installation: nvcc --version

Check that PyTorch can see your GPU:

import torch; print(torch.cuda.is_available())

Getting Help¶

If you continue to have installation problems, please:

Check our GitHub issues
Join our community forum