TensorLy was created in 2015 by Jean Kossaifi to make tensor methods accessible and easy. It was first presented at the NeurIPS workshop on “Learning with Tensors: Why Now and How?” and later published as a JMLR paper titled “TensorLy: Tensor Learning in Python”, by Jean Kossaifi, Yannis Panagakis, Anima Anandkumar and Maja Pantic.
Originally, TensorLy was built on top of NumPy and SciPy only. In order to combine tensor methods with deep learning and run them on multiple devices, CPU and GPU, a flexible backend system was added. This allows algorithms written in TensorLy to be ran with any major framework such as PyTorch, MXNet, TensorFlow, CuPy and JAX.
TensorLy is first and formost a community aiming to make tensor learning easy and accessible.
With a robust and active group of contributors, we would like to thank all those who have contributed, including (alphabetical order):
For a full list of contributors check the Github page.
The TensorLy project is and has been supported by various organizations and universities:
INRIA is funding a full-time engineer to work on TensorLy.