DingXiaoH RepVGG issues

DingXiaoH / RepVGG

RepVGG: Making VGG-style ConvNets Great Again

MIT License

3.34k stars 435 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

question about the newest work?

#122 CoinCheung opened 1 month ago
0
请问RepVGGPlus中的辅助分类器应该如何使用？

#121 BUG423 opened 1 month ago
0
预训练模型是在IN-1k上的吗，有22k上的预训练模型吗？

#120 McCarthyjfq opened 4 months ago
1
Why would you use 3x3+1x1+identity instead of 3x3+3x3+identity?

#119 hopef opened 10 months ago
0
Running example_pspnet.py generates an error: raise NotImplementedError

#118 fhf526 closed 11 months ago
1
The diff is large when BN use other initialization method in verify script

#117 2h4dl opened 1 year ago
0
Question about training and inference models

#116 hosytuyen opened 1 year ago
0
为什么SEBlock中使用Conv2D而不是Linear层？

#115 Z-Xiong closed 1 year ago
1
关于量化的问题，一般是对重参数化后的模型，再进一步进行量化的吗

#114 LKAMING97 opened 1 year ago
0
多分支存在非线性的合并

#113 piaohe20221128 opened 1 year ago
1
Question about RepVGG usage for downstream task

#112 imdsafi09 closed 1 year ago
0
为什么RepVgg的单个block拿出来测试，重参数化前后输出的结果是不一样的

#111 lokiwizard closed 1 year ago
0
a mistake in code

#110 Halance closed 4 months ago
1
jit.trace error

#109 nistarlwc opened 1 year ago
0
关于多卡训练卡死的问题

#108 SEUvictor opened 1 year ago
1
The bug of QAT

#107 qhy991 opened 1 year ago
0
Training script inquiry

#106 zw615 opened 1 year ago
0
Question about multi-branch and single-branch network

#105 MiaoJieF opened 2 years ago
1
subprocess.CalledProcessError: Command '['/usr/bin/python3', '-u', 'main.py', '--local_rank=0']' returned non-zero exit status 1.

#104 starsky68 closed 2 years ago
0
请问如果我用RepVGG作为backbone，并且在neck中有其他的操作，比如5*5卷积，这样训练出来的模型还能做结构重参数化吗？

#103 AssassinJay closed 2 years ago
1
how to prune repvgg

#102 sinoaidi opened 2 years ago
0
Paper Question - Why less favored than MobileNets for low-powered devices?

#101 ksachdeva opened 2 years ago
1
why get different result after reparam

#100 fenghuoxiguozu opened 2 years ago
3
eq_kernel in get_custom_L2 missing contribution of rbr_identity.

#99 vchiley opened 2 years ago
0
question about insert BN before QAT

#98 zhaoxin111 opened 2 years ago
0
How is 1x1 conv with identity matrix as kernel equal to Identity layer

#97 prakashjayy closed 2 years ago
1
mmsegmentation

#96 endaoguansanlu opened 2 years ago
0
convert.py

#95 HonestyBrave closed 2 years ago
0
CIFAR10训练的参数？

#94 wooozihui opened 2 years ago
0
关于repconv switch-to-deploy的问题

#93 pierrelzw opened 2 years ago
0
what matter does "get_custom_L2" do?

#92 shoutOutYangJie opened 2 years ago
1
Re-params quantized RepVGG model

#91 thancaocuong closed 2 years ago
2
Any reference about get_custom_L2?

#90 JoJoistheBestOne closed 2 years ago
1
Paper Question: Use compression idea and generate same result as multi-branch model?

#89 oconnor127 opened 2 years ago
1
why not use python3 -m torch.distributed.launch

#88 xiangyun-hz closed 2 years ago
0
3x3 convolution

#87 SpiritAshes closed 2 years ago
1
Hyperparameters for RepVGGplus-L2pse

#86 zhreshold opened 2 years ago
0
Some questions about quantization

#85 Desmond-97 opened 2 years ago
9
MegEngine Implemention of RepVGG

#84 Asthestarsfalll closed 2 years ago
0
SSD with RepVGG backbone

#83 thuangb opened 2 years ago
2
for style transfer task training

#82 MaltoseMai opened 2 years ago
0
Training with another dataset?

#81 Enescigdem closed 2 years ago
1
questions about training on custom datasets？

#80 Hezhexi2002 opened 3 years ago
0
PTQ精度问题

#79 huoshuai-dot closed 3 years ago
1
How to understand the Structural re-parameterization of a RepVGG block

#78 jyang68sh closed 2 years ago
2
Given input size: (64x240x320). Calculated output size: (64x0x1). Output size is too small

#77 DonMuv opened 3 years ago
1
anyone use repvgg backbone in insightface face recognition task?

#76 xiakj opened 3 years ago
1
Update Arxiv Paper with RepVGGplus

#75 Zacharias030 opened 3 years ago
2
Where is the re-parameter module?

#74 typhoonlee opened 3 years ago
2
Weight decay will be applied to the weight of BN in identity branch and the weight of FC when using --custwd?

#73 wangjinwei94 closed 3 years ago
2