统计predictor的overhead

[guanchenl]

Prerequisites

Before submitting your question, please ensure the following:

• I am running the latest version of PowerInfer. Development is rapid, and as of now, there are no tagged versions.
• I have carefully read and followed the instructions in the README.md.
• I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).

Question Details

有没有什么快速的办法能够统计到predictor的耗时呢，例如输出如下结果：
...
llama_print_timings: eval time = 2082.97 ms / 127 runs ( 16.40 ms per token, 60.97 tokens per second)
llama_print_timings: predictor eval time = xx.xx ms / 127 runs ( xx.xx ms per token, xx.xx tokens per second)
...