aclmdlSetConfigOpt

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 Usage

Set the value of an attribute in the configuration object for loading a model, including the model execution priority, model file directory or memory address, and memory size.

This API needs to work with the following APIs to load models:

Call aclmdlCreateConfigHandle to create a configuration object for loading a model.
(Optional) Call aclmdlSetExternalWeightAddress to configure device memory for storing external weights.
Call aclmdlSetConfigOpt for multiple times to set the value of each attribute in the configuration object.
Call aclmdlLoadWithConfig to specify the configuration information for model loading and load your model.
After the model is successfully loaded, call aclmdlDestroyConfigHandle to destroy the configuration object.

Prototype

aclError aclmdlSetConfigOpt(aclmdlConfigHandle *handle, aclmdlConfigAttr attr, const void *attrValue, size_t valueSize)

Parameters

Parameter	Input/Output	Description
handle	Output	Pointer to the configuration object for model loading. For details about the type definition, see aclmdlConfigHandle. Call aclmdlCreateConfigHandle to create an object in advance.
attr	Input	Attribute to set. For details about the type definition, see aclmdlConfigAttr.
attrValue	Input	Pointer to the attr value to set. If the attribute value itself is a pointer, pass the address of the pointer.
valueSize	Input	Data length of attrValue. You can use the *sizeof(attrValue)** function in the C/C++ standard library to query the data length.

Returns

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

Reference

For model loading by using aclmdlSetConfigOpt and aclmdlLoadWithConfig, the caller needs to set the attributes in the configuration object passed to the API call to decide how the model will be loaded and who will manage the memory.

By choosing from the following APIs, you can directly determine whether to load a model from a file or from memory and whether the memory is managed by the system or the user:

Parent topic: Model Loading and Unloading