Mask

python3 mask_generate_model.py -token <max_token_cnt> -p <process_id> -pool <pool_size> -t <npu_type>

<max_token_cnt>: maximum value of the token generated by the operator. The default value is 2500, and the recommended value range is [1, 300000].

<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 16.

<npu_type>: hardware type. Only 310P is supported.

--help | -h: help information.

N/A