Composable by default
Metrics, datasets, and uncertainty estimates work together without requiring custom pipelines.
Software · Library
torch_measure
A PyTorch library for measurement science. Includes IRT models (Rasch, 2PL, 3PL), computerized adaptive testing, psychometric metrics, and GPU-accelerated estimation.
Design principles
Built for careful measurement.
Three commitments shape every part of the library.
Measurement-aware outputs
Every output makes its assumptions visible. Results always include uncertainty and comparability information.
Built for community use
Researchers across institutions can adopt, extend, and contribute back to shared infrastructure.