aclmdlRIExecuteAsync

Note: This API is for trial use and may be changed in later versions. It is not available in commercial products.

Product	Supported
Atlas 350 Accelerator Card	√
Atlas A3 training product / Atlas A3 inference product	√
Atlas A2 training product / Atlas A2 inference product	√
Atlas 200I/500 A2 inference product	☓
Atlas inference product	√
Atlas training product	☓

Executes the model. This API is asynchronous.

      
           aclError aclmdlRIExecuteAsync(aclmdlRI modelRI, aclrtStream stream)

Parameter	Input/Output	Description
modelRI	Input	Model running instance. For details about the type definition, see aclmdlRI. Obtain a modelRI in either of the following ways: After aclmdlRICaptureBegin is called to capture the tasks delivered to a stream, you can call aclmdlRICaptureGetInfo to obtain the running instance of the model and then pass the instance to the current API. Call APIs such as aclmdlRIBuildBegin and aclmdlRIBuildEnd to build a model running instance, and then pass the instance to the current API.
stream	Input	Stream for executing the model inference task. For details about the type definition, see aclrtStream.

Parameter

Input/Output

Description

modelRI

Input

Model running instance. For details about the type definition, see aclmdlRI.

Obtain a modelRI in either of the following ways:

After aclmdlRICaptureBegin is called to capture the tasks delivered to a stream, you can call aclmdlRICaptureGetInfo to obtain the running instance of the model and then pass the instance to the current API.
Call APIs such as aclmdlRIBuildBegin and aclmdlRIBuildEnd to build a model running instance, and then pass the instance to the current API.

stream

Input

Stream for executing the model inference task. For details about the type definition, see aclrtStream.

0 on success; otherwise, failure. For details, see aclError.

For the API call example, see Single-Stream Capture.

Parent topic: Model Runtime Instance Management