[object Object]

[object Object][object Object]undefined[object Object]

Adds the value in the source tensor to the value of the corresponding position in the input tensor based on the given index in a specified dimension.

[object Object]

Each operator has calls. First, [object Object] is called to obtain the workspace size required for computation and the executor that contains the operator computation flow. Then, [object Object] is called to perform computation.

[object Object]

• Parameters

  [object Object]

• Returns

  [object Object]: status code. For details, see .

  The first-phase API implements input parameter validation. The following error codes may be returned.

  [object Object]

[object Object]

• Parameters

  [object Object]

• Returns

  [object Object]: status code. For details, see .

[object Object]

• Deterministic computation:
  • By default, aclnnIndexAddV2 is implemented in non-deterministic mode. You can enable deterministic computing by using aclrtCtxSetSysParamOpt.
• Value range of index
  • The value of index must be within the range of [0, self.shape[dim]). That is, the index value must be within the shape size of self in the dim dimension. Negative indexes and out-of-bounds indexes are not supported.
• Additional restrictions in high-performance mode (mode = 0):
  • The value of dim is 0 or -2.
  • The data type of self is FLOAT, FLOAT16, INT32, INT16, or BFLOAT16.
  • self is a two-dimensional tensor.
  • If mode is set to 0 and the restrictions in the parameter description are met but the preceding three restrictions are not met, the API throws an error.

[object Object]

The following example is for reference only. For details, see .