runeversion Documentation on ocaml.org
Functional transformations for Nx arrays
Automatic differentiation and vectorizing maps for Nx tensors. Reverse-mode AD (grad, vjp), forward-mode AD (jvp), vmap, and gradient checking, built on OCaml 5 effect handlers.
| Tags | automatic-differentiation machine-learning deep-learning optimization |
|---|---|
| Author | Thibaut Mattio <thibaut.mattio@gmail.com> |
| License | ISC |
| Published | |
| Homepage | https://github.com/raven-ml/raven |
| Issue Tracker | https://github.com/raven-ml/raven/issues |
| Maintainer | Thibaut Mattio <thibaut.mattio@gmail.com> |
| Dependencies | |
| Source [http] | https://github.com/raven-ml/raven/releases/download/1.0.0_alpha3/raven-1.0.0.alpha3.tbz sha256=96d35ce03dfbebd2313657273e24c2e2d20f9e6c7825b8518b69bd1d6ed5870f sha512=90c5053731d4108f37c19430e45456063e872b04b8a1bbad064c356e1b18e69222de8bfcf4ec14757e71f18164ec6e4630ba770dbcb1291665de5418827d1465 |
| Edit | https://github.com/ocaml/opam-repository/tree/master/packages/rune/rune.1.0.0~alpha3/opam |
