FlatInt8AT

python3 flat_at_int8_generate_model.py --cores <core_num> -d <dim> -c <code_num> -p <process_id> --soc-version <soc_version>

<core_num>: number of AI Cores of the Ascend AI Processor. The default value is 2.

<dim>: dimension of the input feature vector.

<code_num>: number of base library features to be compared with the input features.

<process_id>: ID of the process for batch generating operator multi-process scheduling tasks.

<soc_version>: Ascend AI Processor model. The default value is Ascend310P3.

Run the command to obtain a group of operator model files.

FlatInt8AT reduces the time required by train, add, and update in the Atlas inference products.

• dim ∈ {256}
• code_num ∈ {1024, 2048, 4096, 8192, 16384, 32768}
• soc_version ∈ {Ascend310P3}