search

dog-qiuqiu / FastestDet

:zap: A newly designed ultra lightweight anchor free target detection algorithm， weight only 250K parameters， reduces the time consumption by 10% compared with yolo-fastest, and the post-processing is simpler
BSD 3-Clause "New" or "Revised" License
776 stars 133 forks source link

能否有预训练模型? #9

Closed superbayes closed 2 years ago

superbayes commented 2 years ago

能否像yoloV5那样,提供预训练模型(加快训练速度)?

dog-qiuqiu commented 2 years ago

默认就加载预训练模型阿，详细看shufflenet.py

sudospike commented 2 years ago

@dog-qiuqiu shufflenet.py里的backbone权重是什么上面的，是imagenet的权重吗？

Hrayo712 commented 1 year ago

@sudospike did you figure out where the weights come from ?

superbayes commented 1 year ago

应该是pytorch官方的

---- 回复的原邮件 ---- | 发件人 | @.> | | 日期 | 2022年11月04日 18:50 | | 收件人 | @.> | | 抄送至 | @.>@.> | | 主题 | Re: [dog-qiuqiu/FastestDet] 能否有预训练模型? (Issue #9) |

@dog-qiuqiu shufflenet.py里的backbone权重是什么上面的，是imagenet的权重吗？

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>