madgrad

Here are 4 public repositories matching this topic...

optimizer & lr scheduler & loss function collections in PyTorch

deep-learning sam optimizer pytorch ranger muon loss-functions lookahead scion learning-rate-scheduling radam diffgrad gradient-centralization adamp adabelief madgrad adan adai ademamix

Implementation of our paper "Wasserstein Adversarial Transformer for Cloud Workload Prediction"

Super-Convergence on CIFAR10

Provide a collection of flexible learning rate schedulers for PyTorch with presets, warmup wrappers, and research-based implementations.

To associate your repository with the madgrad topic, visit your repo's landing page and select "manage topics."