- Function: Quantizes the input tensor x.
- Formula:
Each operator has calls. First, aclnnQuantizeGetWorkspaceSize is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, aclnnQuantize is called to perform computation.
[object Object]
[object Object]
Parameters
[object Object][object Object]Atlas inference products[object Object]:
- Data type:
- The input parameters
[object Object]and[object Object]do not support BFLOAT16 or FLOAT32. - The input parameter
[object Object]does not support FLOAT32. When the data type is BFLOAT16, the data types of both[object Object]and[object Object]are BFLOAT16. - The output parameter
[object Object]supports only INT8, UINT8, and INT32.
- The input parameters
- The input parameter
[object Object]supports only ACL_INT8, ACL_UINT8, and ACL_INT32.
- Data type:
[object Object]Atlas A2 training products/Atlas A2 inference products[object Object] and [object Object]Atlas A3 training products/Atlas A3 inference products[object Object]:
- Data type:
- The input parameter <idp:inline displayname="code" id="code184913122319">zeroPoints[object Object] does not support FLOAT32. When the data type is BFLOAT16, the data types of both
[object Object]and[object Object]are BFLOAT16. - The output parameter
[object Object]supports only INT8, UINT8, and INT32.
- The input parameter <idp:inline displayname="code" id="code184913122319">zeroPoints[object Object] does not support FLOAT32. When the data type is BFLOAT16, the data types of both
- The input parameter
[object Object]supports only ACL_INT8, ACL_UINT8, and ACL_INT32.
- Data type:
Returns
[object Object]: status code. For details, see .The first-phase API implements input parameter verification. The following errors may be thrown:
[object Object]
- Deterministic compute:
- aclnnQuantize defaults to a deterministic implementation.
The following example is for reference only. For details, see .
[object Object]