Addcdiv
English | 简体中文
Product Support Status
| Product | Supported |
|---|---|
| Ascend 950PR&950DT products | √ |
| Atlas A3 products | √ |
| Atlas A2 products | √ |
| Atlas 200I/500 A2 inference products | × |
| Atlas inference products | × |
| Atlas training products | √ |
Function Description
-
Operator function: a tensor computation function that performs the combined multiply-divide-add operation and merges tensor division (with scaling) + tensor addition into a single operation.
-
Computation formula:
out=self+value×(input1/input2)out = self + value × (input_1 / input_2)
Parameter Description
| Parameter Name | Input/Output/Attribute | Description | Data Type | Data Format |
|---|---|---|---|---|
| self | Input | Input parameter for the addcdiv computation, which is self in the formula. | FLOAT16, BFLOAT16, FLOAT | ND |
| input_1 | Input | Input parameter for the addcdiv computation, which is input_1 in the formula. | FLOAT16, BFLOAT16, FLOAT | ND |
| input_2 | Input | Input parameter for the addcdiv computation, which is input_2 in the formula. | FLOAT16, BFLOAT16, FLOAT | ND |
| value | Input | Input parameter for the addcdiv computation, which is value in the formula. | FLOAT16, BFLOAT16, FLOAT | ND |
| out | Output | Output parameter for the addcdiv computation, which is out in the formula. | FLOAT16, BFLOAT16, FLOAT | ND |
- Atlas training products and Atlas inference products: BFLOAT16 is not supported.
Constraint Description
None
Invocation Description
| Invocation Method | Invocation Sample | Description |
|---|---|---|
| aclnn invocation | test_aclnn_addcdiv | Invokes the Addcdiv operator through the aclnnAddcdiv API. |
| Graph-mode invocation | - | Invokes the Addcdiv operator by building a graph through the operator IR. |