Method InstanceNormalization
InstanceNormalization(TensorFloat, TensorFloat, TensorFloat, float)
Computes the mean variance on the spatial dimensions of the input tensor and normalizes them according to scale and bias tensors.
Declaration
public TensorFloat InstanceNormalization(TensorFloat X, TensorFloat S, TensorFloat B, float epsilon)Parameters
| Type | Name | Description |
|---|---|---|
| TensorFloat | X | The input tensor. |
| TensorFloat | S | The scale tensor. |
| TensorFloat | B | The bias tensor. |
| float | epsilon | The epsilon value the layer uses to avoid division by zero. |
Returns
| Type | Description |
|---|---|
| TensorFloat | The computed output tensor. |