IVFFLAT

python3 ivfflat_generate_model.py -d <dim> -c <coarse_centroid_num> --cores <core_num> -p <process_id> -pool <pool_size> -t <npu_type>

<dim>: feature vector dimension. The default value is 128.

<coarse_centroid_num>: number of level-1 cluster centroids. The default value is 16384.

<core_num>: number of AI Cores of the Ascend AI Processor. The default value is 40. The value of this parameter is determined by <npu_type>. When <npu_type> is set to 910B3, <core_num> is set to 40.

<process_id>: ID of the process for multi-process scheduling of operators generated in batches. The default value is 0, and you do not need to set this parameter.

<pool_size>: size of the process pool for multi-process scheduling of operators generated in batches. The default value is 10.

<npu_type>: hardware form. Currently, it supports the Atlas A2 inference products and Atlas A3 inference products. The default value is 910B4. If you cannot determine the value of npu_type, run the npu-smi info command on the server where the Ascend AI Processor is installed. Delete the last digit of the queried value of Name, and the remaining part is the value of npu_type. For the Atlas 800I A3 SuperPoD Server, run the npu-smi info -t board -i 0 -c 0 command to obtain the NPU Name information. 910_NPU Name is the value of npu_type.

--help | -h: help information.

Run the command to obtain a group of operator model files. You need to modify the parameters in the command.

• dim ∈ {64, 128, 256, 384, 512}
• <coarse_centroid_num> ∈ {1024, 2048, 4096, 8192, 16384, 32768}
• 0 ≤ <pool_size> ≤ 32