aclmdlRIBuildBegin

Applicability

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	√

Function

Starts building a model runtime instance.

After aclmdlRIBuildBegin is called, call aclmdlRIBindStream to bind the model runtime instance to a stream, and then dispatch tasks to the stream. After all tasks are dispatched, call aclmdlRIEndTask to mark the end of task dispatch on the stream, and then call aclmdlRIBuildEnd to end model building. In this case, all tasks dispatched to the stream are not executed immediately. They are executed only when aclmdlRIExecute or aclmdlRIExecuteAsync is called to run model inference.

After all tasks are executed, if the model runtime instance is no longer used, call aclmdlRIUnbindStream to unbind it from the stream. If necessary, call aclmdlRIDestroy to destroy the resource.

Prototype

aclError aclmdlRIBuildBegin(aclmdlRI *modelRI, uint32_t flag)

Parameters

Parameter	Input/Output	Description
modelRI	Input	Model runtime instance, whose model is used to temporarily store the built tasks. For the type definition, see aclmdlRI.
flag	Input	Reserved parameter. The value is fixed at 0.

Returns

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

Parent topic: Model Runtime Instance Management