search

positive666 / yolo_research

based on yolo-high-level project (detect\pose\classify\segment\):include yolov5\yolov7\yolov8\ core ,improvement research ,SwintransformV2 and Attention Series. training skills, business customization, engineering deployment C
GNU General Public License v3.0
758 stars 146 forks source link

BoT #12

Closed Sa-UpWorld closed 2 years ago

Sa-UpWorld commented 2 years ago

❔Question

感觉BoT好像不如C3块和C3TR块，可以减少参数量但会拉低一部分精度啊

Additional context

positive666 commented 2 years ago

❔Question

感觉BoT好像不如C3块和C3TR块，可以减少参数量但会拉低一部分精度啊

Additional context

GAM还可以 你试试 BOT最近没管了 想趁业余时间把SWINV2.0 完善了 不然过一阵子官方都开源了

Sa-UpWorld commented 2 years ago

❔Question

感觉BoT好像不如C3块和C3TR块，可以减少参数量但会拉低一部分精度啊

Additional context

GAM还可以 你试试 BOT最近没管了 想趁业余时间把SWINV2.0 完善了 不然过一阵子官方都开源了

您好，我观察到GAM计算量很大，请问有哪些策略可以降低其复杂性呢，我用了4个GAM增加了20GFLOPs

positive666 commented 2 years ago

❔Question

感觉BoT好像不如C3块和C3TR块，可以减少参数量但会拉低一部分精度啊

Additional context

GAM还可以 你试试 BOT最近没管了 想趁业余时间把SWINV2.0 完善了 不然过一阵子官方都开源了

您好，我观察到GAM计算量很大，请问有哪些策略可以降低其复杂性呢，我用了4个GAM增加了20GFLOPs

正在改 用dwconv 但是测试好像有点问题

github-actions[bot] commented 2 years ago

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.