Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / xiaolai-sqlai/mobilenetv3 issues and pull requests

#51 - SE模块位置与论文中不符？

Issue - State: closed - Opened by Yangworld4 4 months ago

#50 - SeModule Error

Issue - State: open - Opened by wyyaa123 11 months ago - 1 comment

#46 - ！！！

Issue - State: open - Opened by djangaoajb over 1 year ago

#44 - About the Residual Block

Issue - State: open - Opened by enemy1205 about 2 years ago

#43 - added MIT LICENSE

Pull Request - State: open - Opened by mfournarakis about 2 years ago

#42 - 模型训练的配置参数

Issue - State: open - Opened by clxia12 about 3 years ago

#40 - 大模型的与训练权重出来了么

Issue - State: open - Opened by hx121071 almost 4 years ago - 1 comment

#39 - 疑问求教篇

Issue - State: open - Opened by yukaizhou almost 4 years ago

#38 - what's different new and old?

Issue - State: open - Opened by lucasjinreal about 4 years ago

#37 - SeModule

Issue - State: open - Opened by jun214384468 about 4 years ago

#36 - A question about mbv3_small.pth.tar

Issue - State: open - Opened by YuQi9797 about 4 years ago - 1 comment

#35 - What Shortcut will be, if Stride is 2?

Issue - State: open - Opened by BlueAnthony about 4 years ago

#34 - about the model of mobilenetv3_large

Issue - State: open - Opened by freefishman over 4 years ago

#33 - About SE Reduction ratio?

Issue - State: open - Opened by H-YunHui over 4 years ago - 1 comment

#32 - dropout of 0.8 in mobilenetv3 paper

Issue - State: open - Opened by chunhuizng over 4 years ago

#31 - How can I train my own dataset?

Issue - State: open - Opened by YuzhouPeng over 4 years ago

#30 - MobilenetV3 block is totally wrong!

Issue - State: closed - Opened by real-zhangzhe over 4 years ago

#29 - fail to load pre-model

Issue - State: open - Opened by mk123qwe over 4 years ago - 1 comment

#28 - Difference on NBN from paper

Issue - State: closed - Opened by real-zhangzhe over 4 years ago

#27 - Add large version which is same as the structure described in paper.

Pull Request - State: closed - Opened by TKONIY almost 5 years ago

#26 - SE's position differs from paper.

Issue - State: open - Opened by TKONIY almost 5 years ago - 4 comments

#24 - the model of mobilenetv3_large

Issue - State: open - Opened by WangTianYuan almost 5 years ago

#22 - loss nan

Issue - State: open - Opened by jlcai5 about 5 years ago - 1 comment

#21 - 关于评价函数Madds和params可以开源么

Issue - State: open - Opened by Peterisfar about 5 years ago - 1 comment

#20 - Mobile-Net

Issue - State: open - Opened by chuliuT about 5 years ago - 2 comments

#19 - Can anyone really reproduce the result with the code?

Issue - State: open - Opened by triangleCZH about 5 years ago - 7 comments

#18 - Doubt the accuracy

Issue - State: open - Opened by SteveTsui about 5 years ago

#17 - Stride error

Issue - State: open - Opened by SteveTsui over 5 years ago

#16 - Training Code supplement

Issue - State: open - Opened by zengqixun123 over 5 years ago

#15 - Difference on SE from paper

Issue - State: open - Opened by triangleCZH over 5 years ago - 1 comment

#14 - learning rate setting and data augumentation

Issue - State: open - Opened by zhoudaquan over 5 years ago

#13 - MobileNetV3_large network has been updated in the paper

Issue - State: open - Opened by Lukzin over 5 years ago - 1 comment

#12 - The SE module is wrong, it's not just squeeze and expand the channel.

Issue - State: open - Opened by longxianlei over 5 years ago - 1 comment

#11 - The last two FC layers don't need batch norm.

Issue - State: open - Opened by longxianlei over 5 years ago - 1 comment

#10 - 模型解压报错

Issue - State: open - Opened by smartadpole over 5 years ago - 1 comment

#8 - Validation accuracy for large model low, mistakes in model

Issue - State: open - Opened by rwightman over 5 years ago - 8 comments

#7 - AdaptiveAvgPool2d seems not used in compute graph

Issue - State: closed - Opened by ghost over 5 years ago - 1 comment

#6 - pre-trained weight decompression error

Issue - State: closed - Opened by feiyun1265 over 5 years ago

#5 - Validation accuracy for mobilenetv3_small

Issue - State: open - Opened by sjzhuo over 5 years ago - 3 comments

#4 - Why do you use conv and bn in SeModule?

Issue - State: open - Opened by kuan-wang over 5 years ago - 3 comments

#3 - Why you didn't use avg pool in SeModule?

Issue - State: closed - Opened by PistonY over 5 years ago - 5 comments

#2 - batchnorm handing issue when inference

Issue - State: closed - Opened by lucasjinreal over 5 years ago - 4 comments

#1 - Image preprocessing method

Issue - State: open - Opened by zhangming8 over 5 years ago