Installation¶
Basic install¶
pip install trnsolver
With Neuron hardware support¶
On a Trainium/Inferentia instance, install into the AMI's pre-built Neuron venv (which already contains neuronxcc, since it's not on public PyPI):
source /opt/aws_neuronx_venv_pytorch_2_9/bin/activate
pip install trnsolver[dev]
The trnsolver[neuron] extra is only useful when building a custom Neuron environment from scratch.
Development install¶
git clone https://github.com/trnsci/trnsolver.git
cd trnsolver
pip install -e ".[dev]"
pytest tests/ -v
Requirements¶
- Python >= 3.10
- PyTorch >= 2.1
- NumPy >= 1.24
- neuronxcc >= 2.15 (optional, for Trainium hardware)
- torch-neuronx >= 2.1 (optional, for Trainium hardware)
Hardware compatibility¶
The NKI Jacobi rotation kernel is scaffolded but not yet validated on hardware — trnsolver.eigh falls back to torch.linalg.eigh whenever the NKI path isn't selected. CPU/PyTorch correctness is preserved across all platforms.
For a pre-built CI instance, see AWS Setup and the Terraform module in infra/terraform/.