napistu-torch 0.4.0

PyTorch-based toolkit for working with Napistu network graphs

Napistu PyTorch Library

Python package supporting integration of Napistu network graphs and PyTorch NNs

This Python package builds on the Napistu Python package with PyTorch-specific data, model, and results management. As part of the broader Napistu project this work is intended to create and interrogate genome-scale networks of cellular physiology.

Setup

Napistu is available on PyPI so the recommended way to use it is just to pip install with:

pip install napistu-torch

Alternatively, you can clone this repository and perform a local install. e.g., from this directory:

pip install .

Mac setup

uv pip install torch==2.8.0
uv pip install torch-scatter torch-sparse -f https://data.pyg.org/whl/torch-2.8.0+cpu.html
uv install 'napistu-torch[dev]'

Documentation

• 🚸 Project Documentation: napistu/wiki

Advanced setup

Wandb

If this is your first time using wandb, you'll need to authenticate:

1. Go to https://wandb.ai/ and create an account
2. Get your API key from https://wandb.ai/authorize
3. Run: wandb login
4. Or set environment variable: export WANDB_API_KEY=your_api_key_here