search

Lam1360 / YOLOv3-model-pruning

在 oxford hand 数据集上对 YOLOv3 做模型剪枝（network slimming）
MIT License
1.67k stars 432 forks source link

保存的.weights模型为啥ｄａｒｋｎｅｔ调用无法检测 #75

Open lilei1128 opened 4 years ago

lilei1128 commented 4 years ago

用ｍｏｄｅｌ中的ｓａｖｅ_darknet_weights（）函数保存的.weights格式模型，为啥在ｄａｒｋｎｅｔ中调用检测不到目标？？？

wangruo91 commented 4 years ago

@lilei1128 Hi，我遇到和你一样的问题，你后来解决了么？

lilei1128 commented 4 years ago

一个网友说可以把阈值设置的低一点可以检测，你试一下。

------------------ 原始邮件 ------------------ 发件人: "wangruo91"<notifications@github.com>; 发送时间: 2020年2月10日(星期一) 凌晨0:52 收件人: "Lam1360/YOLOv3-model-pruning"<YOLOv3-model-pruning@noreply.github.com>; 抄送: "LeL"<1715734693@qq.com>;"Mention"<mention@noreply.github.com>; 主题: Re: [Lam1360/YOLOv3-model-pruning] 保存的.weights模型为啥ｄａｒｋｎｅｔ调用无法检测 (#75)

@lilei1128 Hi，我遇到和你一样的问题，你后来解决了么？

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

wangruo91 commented 4 years ago

@lilei1128 -thresh设成0.1了都没用，决定直接修改darknet做稀疏训练了。

Kitty-Learn commented 4 years ago

You can use darknet to retrain pruned-model

lilei1128 commented 4 years ago

You can use darknet to retrain pruned-model

yeah, I thought of this method but didnt try it.