v2.2.0

New features:

• (#48) Intermediates can now be shared between contractions, see here for more details.
• (#53) Intermediate caching is thread safe.

Enhancements:

• (#48) Expressions are now mapped to non-unicode index set so that unicode input is support for all backends.
• (#58) Adds tensorflow and theano with shared intermediates.

Bug fixes:

• (#41) PyTorch indices are mapped back to a small a-z subset valid for PyTorch's einsum implementation.

v2.1.3

Bug fixes:

• Fixes unicode issue for large numbers of tensors in Python 2.7.
• Fixes unicode install bug in README.md.

v2.1.2

Bug Fixes:

• Ensures versioneer.py is in MANIFEST.in for a clean pip install.

v2.1.1

Bug Fixes:

• Minor tweak to release procedure.

v2.1.0

opt_einsum continues to improve its support for additional backends beyond NumPy with PyTorch.

We have also published the opt_einsum package in the Journal of Open Source Software. If you use this package in your work, please consider citing us!

New features:

• PyTorch backend support
• Tensorflow eager-mode execution backend support

Enhancements:

• Intermediate tensordot-like expressions are now ordered to avoid transposes.
• CI now uses conda backend to better support GPU and tensor libraries.
• Now accepts arbitrary unicode indices rather than a subset.
• New auto path option which switches between optimal and greedy at four tensors.

Bug fixes:

• Fixed issue where broadcast indices were incorrectly locked out of tensordot-like evaluations even after their dimension was broadcast.

v2.0.1

opt_einsum is a powerful tensor contraction order optimizer for NumPy and related ecosystems.

New Features

• Allows unlimited Unicode indices.
• Adds a Journal of Open-Source Software paper.
• Minor documentation improvements.

v2.0.0

opt_einsum is a powerful tensor contraction order optimizer for NumPy and related ecosystems.

New Features

• Expressions can be precompiled so that the expression optimization need not happen multiple times.
• The greedy order optimization algorithm has been tuned to be able to handle hundreds of tensors in several seconds.
• Input indices can now be unicode so that expressions can have many thousands of indices.
• GPU and distributed computing backends have been added such as Dask, TensorFlow, CUPy, Theano, and Sparse.

Bug Fixes

• A error effecting cases where opt_einsum mistook broadcasting operations for matrix multiply has been fixed.
• Most error messages are now more expressive.

v1.0

Official 1.0 release.

Einsum is a very powerful function for contracting tensors of arbitrary dimension and index. However, it is only optimized to contract two terms at a time resulting in non-optimal scaling for contractions with many terms. Opt_einsum aims to fix this by optimizing the contraction order which can lead to arbitrarily large speed ups at the cost of additional intermediate tensors.

Opt_einsum is also implemented into the np.einsum function as of NumPy v1.12.

New BLAS support!

A large step towards to a full 1.0 release. BLAS usage is now automatically applied to all operations. Future releases will be more careful with regard to views and needless data copying.

Python 3 + `setup.py`

Adds Python 3 support in addition to installation through a setup.py command.