Tangent
A Python library for automatic differentiation that generates readable Python source code as its derivative output.
What is Tangent?
Tangent is a Python library for automatic differentiation that generates human-readable Python source code as its derivative output. It performs ahead-of-time autodiff directly on the Python source code, unlike runtime tracing or graph-based approaches, solving the problem of opaque and hard-to-debug derivative computations in machine learning.
Researchers and students in machine learning who want to write models in Python and need to read, debug, or manipulate automatically generated derivative code without sacrificing speed or flexibility.
Developers choose Tangent for its unique source-to-source approach, which produces debuggable and transparent derivative code, supports custom gradients and control flow, and integrates with TensorFlow and NumPy while offering both reverse-mode and forward-mode autodiff.
Overview
Source-to-Source Debuggable Derivatives in Pure Python
Use Cases
Best For
- Debugging automatic derivative code in machine learning models
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
