Function: execute_v2

aclError aclmdlExecuteV2(uint32_t modelId, const aclmdlDataset *input, aclmdlDataset *output, aclrtStream stream, const aclmdlExecConfigHandle *handle)

ret = acl.mdl.execute_v2(model_id, input, output, stream, handle)

Performs model inference based on the attributes configured by the acl.mdl.set_exec_config_opt API until the inference result is returned. This API is enhanced based on acl.mdl.execute. It can control the timeout interval of stream tasks and event tasks during model inference.

model_id: int, ID of the model to be inferred.

You can obtain the model ID after the model is successfully loaded by calling the following APIs:

• acl.mdl.load_from_file
• acl.mdl.load_from_mem
• acl.mdl.load_from_file_with_mem
• acl.mdl.load_from_mem_with_mem

input: int, pointer address of the input data for model inference. For details, see aclmdlDataset.

output: int, pointer address of the output data for model inference. For details, see aclmdlDataset.

When calling acl.create_data_buffer to create an ACL data buffer for storing the output data of the corresponding index, you can set data to 0 and set size to 0 to create an empty ACL data buffer. During model execution, the system calculates and allocates the memory for the index output.

This method saves memory. However, you need to free the memory and reset the aclDataBuffer after using the data. In addition, memory copy is involved when the system allocates memory, which may cause performance loss.

The sample code for freeing the memory and resetting the aclDataBuffer is as follows:

data_buffer = acl.mdl.get_dataset_buffer(output, 0) // Obtain the corresponding data buffer based on the index.
data_addr = acl.get_data_buffer_addr(data_buffer) // Obtain the device pointer address of the data.
acl.rt.free(data_addr) // Release the device memory.
acl.update_data_buffer(data_buffer, 0, 0) // Reset the data buffer for next inference.

stream: int, pointer address of the created stream. To specify a new stream, you can create and obtain the pointer address of the stream by calling acl.rt.create_stream.

handle: int, pointer address of the configuration object executed by the model. The value must be the same as that of handle in acl.mdl.set_exec_config_opt.

ret: int, error code.

• 0 indicates success.
• Other values indicate failure.

• This API must be used together with other APIs to implement model execution. The API call sequence is as follows:
  1. Call acl.mdl.create_exec_config_handle to create the pointer address to the configuration object executed by the model.
  2. Call acl.mdl.set_exec_config_opt for multiple times to set the value of each attribute in the configuration object.
  3. Call acl.mdl.execute_v2 to specify the configuration information required for model execution and execute the model.
  4. Call acl.mdl.destroy_exec_config_handle to destroy the configuration object after the model is successfully loaded.
• Other restrictions are the same as those of acl.mdl.execute.

• For details about the API call sequence and sample code, see Executing a Model.
• For details about the stream creation and management, see Stream Management.