Class RMSNorm
java.lang.Object
ai.beestream.jitnn.nn.Module
ai.beestream.jitnn.nn.RMSNorm
- All Implemented Interfaces:
AutoCloseable
RMSNorm over the last axis, with weight 1 initially, like PyTorch's
nn.RMSNorm.-
Constructor Summary
ConstructorsConstructorDescriptionRMSNorm(int features, float eps) RMSNorm overfeaturesvalues with the given eps. -
Method Summary
Methods inherited from class Module
buffer, buffers, child, close, isClosed, isTraining, loadStateDict, parameter, parameterCount, parameters, stateDict, train, zeroGrad
-
Constructor Details
-
RMSNorm
public RMSNorm(int features, float eps) RMSNorm overfeaturesvalues with the given eps.
-
-
Method Details
-
weight
The scale [features]. -
forward
-