search

mit-han-lab / tinyengine

[NeurIPS 2020] MCUNet: Tiny Deep Learning on IoT Devices; [NeurIPS 2021] MCUNetV2: Memory-Efficient Patch-based Inference for Tiny Deep Learning; [NeurIPS 2022] MCUNetV3: On-Device Training Under 256KB Memory
https://mcunet.mit.edu
MIT License
792 stars 130 forks source link

some api used in "GeneralMemoryScheduler.py" #4

Closed waquey closed 1 year ago

waquey commented 2 years ago

Excuse me, I cannot trace the exact definition of the following api which are related to op

layermem["MAC"] = op.get_macs() layermem["activation"] = op.get_activation_size() layermem["scale"] = op.get_scale_size() layermem["runtime"] = op.get_sbuf_size() layermem["kernel"] = op.get_kbuf_size()

Where can we find the definition of get_macs()/get_activation_size()/get_scale_size()/get_sbuf_size()/get_kbuf_size() ? Could you please help comment that? Thanks

meenchen commented 2 years ago

Hi @waquey, thanks for your interest in our work! Please refer to the implementation of operators in code_generator/operators. The abstraction of these APIs is in code_generator/operators/basic_utils.py.

meenchen commented 1 year ago

Close due to inactivity. Feel free to reopen.