search

THU-MIG / torch-model-compression

针对pytorch模型的自动化模型结构分析和修改工具集，包含自动分析模型结构的模型压缩算法库
MIT License
240 stars 40 forks source link

Does there any params and speed comparasion on common models？ #10

Open lucasjinreal opened 2 years ago

lucasjinreal commented 2 years ago

Hi, does there any speed test on common models such as resnet50, YOLOv4 before and after compression? if there is a table indicates that would be very nice

gdh1995 commented 2 years ago

Um, not yet. Since it varies among devices and flops percentages, it's a bit too hard to write a table for it.

lucasjinreal commented 2 years ago

@gdh1995 i think it's not need speed among different devices, just compare same model on same device, but before and after prune. That's enough convincing, otherwise not so convincing.