Installation

Requirements

Package	Version	Purpose
Python	≥ 3.10	Runtime
NumPy	≥ 1.24	Array operations throughout
SciPy	≥ 1.10	FFT convolution, NNLS, curve fitting, spherical Bessel functions
Matplotlib	≥ 3.7	All plotting functions in qens.plotting
h5py	≥ 3.7	Reading .nxspe HDF5 files from ISIS Mantid
emcee	≥ 3.1 (optional)	Ensemble MCMC sampler. Falls back to Metropolis-Hastings if absent.

From PyPI

pip install qens

# with the recommended ensemble sampler
pip install "qens[mcmc]"

From Source (recommended for development)

git clone https://github.com/ShubhamX57/QENS.git
cd QENS
pip install -e ".[mcmc]"

Verify

import qens
print(qens.__version__)   # 2.0.0
print(qens.available_models())
# ['anisotropic_rotor', 'isotropic_rotor', 'translation_only']

Note

If emcee is not installed, run_mcmc() prints a warning and falls back to a 4-chain Metropolis-Hastings sampler. Results are statistically equivalent but the ensemble autocorrelation time estimate is not available.

Connecting to Read the Docs

The repo ships a .readthedocs.yaml at the project root. To publish:

1. Push your fork to GitHub.

2. Log in at readthedocs.org with your GitHub account.

3. Click Import a Project and select ShubhamX57/QENS.

4. Read the Docs detects .readthedocs.yaml automatically — click Build.

5. Docs go live at https://qens.readthedocs.io/en/latest/.

Every git push to main triggers an automatic rebuild.