Provides a near drop-in capability for performing forward mode automatic differentiation for numpy-based numerical software. For more details on dual-algebra based forward mode automatic differentiation see, for example, this Wikipedia article
- Supports regular numpy arrays as well as arrays with units using Pint and astropy (the latter is less tested).
- Can store multiple Jacobians for an array with respect to different variables
- Jacobians can be stored as full, diagonal (starting point), or sparse matrices (the latter using
scipy.sparsewith occasional use of the (non-scipy)sparsepackage). - Supports most standard operations and functions, but always scope for more to be added
This guide provides a quick way to get started with our project. Please see our [docs]([INSERT LINK TO DOCS SITE / WIKI HERE]) for a more comprehensive overview.
- [INSERT LIST OF REQUIREMENTS HERE]
- [INSERT STEP-BY-STEP SETUP INSTRUCTIONS HERE, WITH OPTIONAL SCREENSHOTS]
- [INSERT STEP-BY-STEP RUN INSTRUCTIONS HERE, WITH OPTIONAL SCREENSHOTS]
- [INSERT LIST OF COMMON USAGE EXAMPLES HERE, WITH OPTIONAL SCREENSHOTS]
- [INSERT STEP-BY-STEP BUILD INSTRUCTIONS HERE, WITH OPTIONAL SCREENSHOTS]
- [INSERT STEP-BY-STEP TEST INSTRUCTIONS HERE, WITH OPTIONAL SCREENSHOTS]
See our CHANGELOG.md for a history of our changes.
See our [releases page]([INSERT LINK TO YOUR RELEASES PAGE]) for our key versioned releases.
[INSERT LINK TO FAQ PAGE OR PROVIDE FAQ INLINE HERE]
[INSERT LINK TO CONTRIBUTING GUIDE OR FILL INLINE HERE]
[INSERT LINK TO YOUR CODE_OF_CONDUCT.md OR SHARE TEXT HERE]
[INSERT LINK TO YOUR GOVERNANCE.md OR SHARE TEXT HERE]
See our: LICENSE
[INSERT CONTACT INFORMATION OR PROFILE LINKS TO MAINTAINERS AMONG COMMITTER LIST]