#149 · Primary category: Deep Learning Frameworks

Theano

Theano was a Python library that allows you to define, optimize, and evaluate mathematical expressions involving multi-dimensional arrays efficiently. It is being continued as PyTensor: www.github.com/pymc-devs/pytensor

Project last updated:01/15/24

GitHub Stars

10.0K

Forks

2.5K

Contributors

386

License

Other

Why we included this project

Theano is a Python library for symbolic computation: you define expressions over multi-dimensional arrays and it compiles them into optimized code that runs on CPU or GPU. It powered a large share of early deep learning research and academic papers, and its approach will feel familiar to anyone who has worked with probabilistic modeling or custom neural network layers. The original maintainers have stopped developing it, but the codebase continues as PyTensor, maintained by the PyMC developers, and that fork still handles modern Bayesian modeling workloads. If you maintain legacy Theano code or want to study the symbolic-graph design that later frameworks borrowed from, this repository is the reference implementation. For new projects, follow the pointer to PyTensor rather than building on the archived code.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category