Visible to Intel only — GUID: GUID-25CDF023-1B62-4A0F-B8EF-0E625C366517
Visible to Intel only — GUID: GUID-25CDF023-1B62-4A0F-B8EF-0E625C366517
LayerNorm
General
LayerNorm performs a layer normalization operation on tensor.
The layerNorm operation performs normalization from begin_norm_axis to last dimension of the data tensor. It is defined by the following formulas which is the same as Layer Normalization.
where
are optional scale and shift for a channel
are mean and variance (see
is a constant to improve numerical stability.
Mean and variance are computed at runtime or provided by a user. When mean and variance are computed at runtime, the following formulas are used:
,
.
Operation attributes
Attribute Name |
Description |
Value Type |
Supported Values |
Required or Optional |
---|---|---|---|---|
Indicate whether to output mean and variance which can be later passed to backward op. |
bool |
false , true (default) |
Optional |
|
begin_norm_axis is used to indicate which axis to start layer normalization. The normalization is from begin_norm_axis to last dimension. Negative values means indexing from right to left. This op normalizes over the last dimension by default, e.g. C in TNC for 3D and LDNC for 4D. |
s64 |
[-r,r-1],where r=rank(src). -1 is default |
Optional |
|
When set to True, this module has learnable per-element affine parameters. |
bool |
false , true (default) |
Optional |
|
The constant to improve numerical stability. |
f32 |
Arbitrary positive f32 value, 1e-5 (default) |
Optional |
Execution arguments
The inputs and outputs must be provided according to below index order when constructing an operation.
Inputs
Index |
Argument Name |
Required or Optional |
---|---|---|
0 |
src |
Required |
1 |
gamma |
Optional |
2 |
beta |
Optional |
Outputs
Index |
Argument Name |
Required or Optional |
---|---|---|
0 |
dst |
Required |
1 |
mean |
Optional |
2 |
variance |
Optional |
Supported data types
LayerNorm operation supports the following data type combinations.
Src / Dst |
Gamma / Beta / Mean / Variance |
---|---|
f32 |
f32 |
bf16 |
f32, bf16 |
f16 |
f32 |