DingXiaoH ACNet issues - Githubissues

DingXiaoH / ACNet

ACNet: Strengthening the Kernel Skeletons for Powerful CNN via Asymmetric Convolution Blocks

MIT License

829 stars 133 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

acb.py中ACBlock类的argument 没有“use_last_bn”

#51 7eCc4 opened 1 month ago
0
请问有TensorFlow1.x的版本吗？

#50 sibangde opened 2 years ago
0
验证时如何把推理转换后的模型变回去

#49 LHH20000923 closed 2 years ago
0
关于ACB的用法

#48 kuazhangxiaoai opened 2 years ago
0
Why use this function, croplayer

#47 X-rap-rose opened 3 years ago
1
can not find the models

#46 JingKunYan1999 closed 3 years ago
2
关于ACBlock部分逻辑的疑问

#45 njustBShen closed 3 years ago
1
About pretrained model

#44 mty62 closed 3 years ago
0
关于acb与conv不完全等价的问题

#43 Orangerccc closed 3 years ago
3
关于ACblock ，border的卷积核构建

#42 DDxk369 closed 3 years ago
1
About params and flops

#41 HCookY95 closed 3 years ago
0
Plug-in version implementation

#40 zjykzj closed 3 years ago
2
depth conv的实验

#39 pawopawo closed 3 years ago
5
how to deploy the trained model ？

#38 luxdong closed 3 years ago
0
请问可以共享一下imagnet训练集的预训练权重吗？

#37 icey-zhang closed 3 years ago
0
weights translation

#36 yanlongbinluck closed 3 years ago
0
About BN layer

#35 HCookY95 closed 3 years ago
1
绘制曲线

#34 ZangHuanyu closed 3 years ago
1
Error in ACBuilder

#33 ZangHuanyu closed 3 years ago
5
Does the val loss match the real performance of the inference model?

#32 SingL3 closed 3 years ago
2
How to use for LRN?

#31 LiuAlex1109 closed 3 years ago
0
请问如何应用到resnet里？

#30 ZangHuanyu closed 3 years ago
9
预训练权重，这个到哪里下载

#29 zhangyu13141 closed 3 years ago
0
train loss and accuracy curve for single gpu

#28 zcc1019 closed 3 years ago
1
about the parameter innitialization of BatchNorm

#27 zcc1019 closed 3 years ago
1
Fail to Produce ResNet 56 （w/o acblock）on cifar 10， on get 94.0x%

#26 LeeJZh closed 3 years ago
3
my data

#25 FengJin-cv closed 3 years ago
0
in the function BNReLUConv2d, is there no BN layer in the ACBlock?

#24 Yannis1995 closed 3 years ago
0
Is there any news about AsymConvBlock?

#23 jianlong-yuan closed 3 years ago
0
Can I directly replace conv2d in CNN with acblock

#22 lofyol closed 4 years ago
0
ResNet56

#21 zhang0391 closed 3 years ago
0
How to use ACNet in DenseNet

#20 zhang0391 closed 3 years ago
0
Is the difference from the original paper only in the learning rate?

#19 pjtmdpj223 closed 3 years ago
1
ACNet reimplement based on PaddlePaddle

#18 parap1uie-s opened 4 years ago
1
Hello, is there any implementation of keras version?

#17 David-19940718 closed 3 years ago
0
How about constructing an ACB with several different dxd kernels?

#16 wangxianliang closed 3 years ago
1
resnet50+acb

#15 anti199994 closed 3 years ago
1
about TODOS

#14 sduzk closed 3 years ago
0
I know why there are so few stars.

#13 Qingfeng0623 closed 3 years ago
2
when will release ImageNet pretrain model?

#11 L1aoXingyu closed 3 years ago
6
TypeError problem

#10 JunMingLu closed 3 years ago
0
Performance if we put the BN after merging the branches

#9 liuzili97 closed 4 years ago
1
Release a PyTorch module

#8 linxueya closed 3 years ago
2
ACNet applies to 3D convolution

#7 weibienao closed 3 years ago
0
what about use dw-conv instead of conv in ACBlock?

#6 zimenglan-sysu-512 closed 4 years ago
1
Good work but more analysis of acnet could be done

#5 xw324 closed 4 years ago
1
question about the behavior of bn layers in inference

#4 CoinCheung closed 4 years ago
4
this work is great, but can you give a example how we can replace 3*3 conv by using 3*3+1*3+3*1

#3 nankeermeng closed 4 years ago
3
ACBlock the diff deploy and train?

#2 ranjiewwen closed 4 years ago
1
Different result?

#1 evi-Genius closed 4 years ago
4