IDetachedTensor<T>
Interfaces · AiDotNet.SelfSupervisedLearning
Marker interface for tensors that should not receive gradients.
Known implementations: DetachedTensor<T>
How It Works
This can be used to mark tensors at compile time as non-differentiable. Useful for type-safe gradient handling in advanced scenarios.
Properties
| Property | Summary |
|---|---|
Tensor<T> Data { get; } |
Gets the underlying tensor data. |