xiaolai-sqlai mobilenetv3 issues

xiaolai-sqlai / mobilenetv3

mobilenetv3 with pytorch，provide pre-train model

MIT License

1.6k stars 340 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

SE模块位置与论文中不符？

#51 Yangworld4 closed 1 month ago
0
SeModule Error

#50 wyyaa123 opened 8 months ago
0
求您训练好的这个MobileNetV3模型较原论文中的模型的改进之处

#49 YangPeng61790312 opened 9 months ago
0
你好，请问params为什么我用 summary测试出来small的6.38m，large为19.76m，请问是什么原因呢？

#48 zclyaya opened 10 months ago
1
增加了使用fcmae自监督训练方法来预训练mobilenetv3网络的代码

#47 yaoyi30 opened 11 months ago
0
！！！

#46 djangaoajb opened 1 year ago
0
这个可以用作人脸识别嘛，需要更改网络结构吗

#45 szgy66 opened 1 year ago
0
About the Residual Block

#44 enemy1205 opened 1 year ago
0
added MIT LICENSE

#43 mfournarakis opened 1 year ago
0
模型训练的配置参数

#42 clxia12 opened 3 years ago
0
分类改为分割，怎么修改，有没有大佬指导下

#41 Songxinlei opened 3 years ago
0
大模型的与训练权重出来了么

#40 hx121071 opened 3 years ago
1
疑问求教篇

#39 yukaizhou opened 3 years ago
0
what's different new and old?

#38 lucasjinreal opened 3 years ago
0
SeModule

#37 jun214384468 opened 3 years ago
0
A question about mbv3_small.pth.tar

#36 YuQi9797 opened 3 years ago
1
What Shortcut will be, if Stride is 2?

#35 BlueAnthony opened 3 years ago
0
about the model of mobilenetv3_large

#34 freefishman opened 4 years ago
0
About SE Reduction ratio?

#33 H-YunHui opened 4 years ago
1
dropout of 0.8 in mobilenetv3 paper

#32 chunhuizng opened 4 years ago
0
How can I train my own dataset?

#31 YuzhouPeng opened 4 years ago
0
MobilenetV3 block is totally wrong!

#30 real-zhangzhe closed 4 years ago
0
fail to load pre-model

#29 mk123qwe opened 4 years ago
1
Difference on NBN from paper

#28 real-zhangzhe closed 4 years ago
0
Add large version which is same as the structure described in paper.

#27 TKONIY closed 4 years ago
0
SE's position differs from paper.

#26 TKONIY opened 4 years ago
4
这是做分类的网络，那检测的呢？怎样修改？

#25 henbucuoshanghai opened 4 years ago
0
the model of mobilenetv3_large

#24 WangTianYuan opened 4 years ago
0
Could you list all of your details that are different from the original paper in README ?

#23 DocF opened 4 years ago
0
loss nan

#22 jlcai5 opened 4 years ago
1
关于评价函数Madds和params可以开源么

#21 Peterisfar opened 4 years ago
1
Mobile-Net

#20 chuliuT opened 4 years ago
2
Can anyone really reproduce the result with the code?

#19 triangleCZH opened 4 years ago
7
Doubt the accuracy

#18 SteveTsui opened 5 years ago
0
Stride error

#17 SteveTsui opened 5 years ago
0
Training Code supplement

#16 zengqixun123 opened 5 years ago
0
Difference on SE from paper

#15 triangleCZH opened 5 years ago
1
learning rate setting and data augumentation

#14 zhoudaquan opened 5 years ago
0
MobileNetV3_large network has been updated in the paper

#13 Lukzin opened 5 years ago
1
The SE module is wrong, it's not just squeeze and expand the channel.

#12 longxianlei opened 5 years ago
1
The last two FC layers don't need batch norm.

#11 longxianlei opened 5 years ago
1
模型解压报错

#10 smartadpole opened 5 years ago
1
the model download can not decompression， can re upload again? thanks a lot , or send to my email ? 328511963@qq.com thanks a lot

#9 cuicuizhang1989 opened 5 years ago
5
Validation accuracy for large model low, mistakes in model

#8 rwightman opened 5 years ago
8
AdaptiveAvgPool2d seems not used in compute graph

#7 ghost closed 5 years ago
1
pre-trained weight decompression error

#6 feiyun1265 closed 5 years ago
0
Validation accuracy for mobilenetv3_small

#5 sjzhuo opened 5 years ago
3
Why do you use conv and bn in SeModule?

#4 kuan-wang opened 5 years ago
3
Why you didn't use avg pool in SeModule?

#3 PistonY closed 5 years ago
5
batchnorm handing issue when inference

#2 lucasjinreal closed 5 years ago
4