Visible to Intel only — GUID: GUID-A910CD8A-F237-4272-8CB9-67CA29D429D8
Visible to Intel only — GUID: GUID-A910CD8A-F237-4272-8CB9-67CA29D429D8
BiasAddBackward
General
BiasAddBackward operation computes the gradients on the bias tensor for BiasAdd operator. This op accumulates all the values from into the channel dimension, the axis depends on the layout of tensor.
Operation attributes
Attribute Name |
Description |
Value Type |
Supported Values |
Required or Optional |
---|---|---|---|---|
Controls how to interpret the shape of diff_dst and diff_bias . |
string |
NCX , NXC (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 |
diff_dst |
Required |
Outputs
Index |
Argument Name |
Required or Optional |
---|---|---|
0 |
diff_bias |
Required |
Supported data types
BiasAddBackward operation supports the following data type combinations.
Diff_dst |
Diff_bias |
---|---|
f32 |
f32 |
bf16 |
bf16 |
f16 |
f16 |