Package: madgrad 0.1.0

madgrad: 'MADGRAD' Method for Stochastic Optimization

A Momentumized, Adaptive, Dual Averaged Gradient Method for Stochastic Optimization algorithm. MADGRAD is a 'best-of-both-worlds' optimizer with the generalization performance of stochastic gradient descent and at least as fast convergence as that of Adam, often faster. A drop-in optim_madgrad() implementation is provided based on Defazio et al (2020) <arxiv:2101.11075>.

Authors:Daniel Falbel [aut, cre, cph], RStudio [cph], MADGRAD original implementation authors. [cph]

madgrad_0.1.0.tar.gz
madgrad_0.1.0.zip(r-4.5)madgrad_0.1.0.zip(r-4.4)madgrad_0.1.0.zip(r-4.3)
madgrad_0.1.0.tgz(r-4.4-any)madgrad_0.1.0.tgz(r-4.3-any)
madgrad_0.1.0.tar.gz(r-4.5-noble)madgrad_0.1.0.tar.gz(r-4.4-noble)
madgrad_0.1.0.tgz(r-4.4-emscripten)madgrad_0.1.0.tgz(r-4.3-emscripten)
madgrad.pdf |madgrad.html✨
madgrad/json (API)

Peer review:

On CRAN:

This package does not link to any Github/Gitlab/R-forge repository. No issue tracker or development information is available.

1 exports 0.00 score 18 dependencies 8 scripts 154 downloads

Last updated 3 years agofrom:40be6a66fe. Checks:OK: 1 NOTE: 6. Indexed: yes.

Exports:optim_madgrad

Dependencies:bitbit64callrclicorodescellipsisgluejsonlitemagrittrprocessxpsR6Rcpprlangsafetensorstorchwithr