Uses of Package
ai.beestream.jitnn.nn.optim
Packages that use ai.beestream.jitnn.nn.optim
Package
Description
Optimisers with PyTorch's semantics (SGD, Adam and AdamW, AdaGrad, RMSProp, AdaDelta, AdaMax,
NAdam, LAMB, LARS), learning-rate schedules (
LRScheduler) and dynamic
loss scaling for FP16 training (GradScaler).-
Classes in ai.beestream.jitnn.nn.optim used by ai.beestream.jitnn.nn.optimClassDescriptionAdam (Kingma and Ba) with L2 weight decay, or AdamW (decoupled decay) via
Adam.adamW(List, float, float); withamsgrad, AMSGrad's running maximum of the second moment (Reddi et al.).Sets an optimiser's learning rate per epoch; callLRScheduler.step()once at the end of each epoch, as with PyTorch's schedulers.Updates parameters in place from their gradients.