Visual-Attention-Network/SegNeXt

Issues

• 大佬们你们下载的哪个版本的mmcv 为什么我版本较老（mmcv-full1.6.0） 会报错 版本太新(mmcv r4)没有mmcv.runner？

  #64 opened 2 months ago by llyyccccc
  1

• Can you provide the IOU per class of IsAID dataset?

  #63 opened 6 months ago by DotWang
  0

• Config file for the i-SAID dataset

  #62 opened 6 months ago by giuseppemartino26
  0

• How to stop shuffle for test.py

  #61 opened 6 months ago by matz23
  0

• Evaluation Result on Cityscapes

  #60 opened 6 months ago by 554773162
  0

• If I want to train with 4 GPUs, do I need to change the code

  #59 opened 7 months ago by alon2U
  0

• The paddlepaddle version of SegNeXt

  #50 opened 9 months ago by Asthestarsfalll
  0

• MACs, not FLOPs

  #49 opened a year ago by 3maoyap
  1

• about AttentionModule

  #58 opened 10 months ago by qiaoqiangPro
  0

• Training gpu consumption vs inference gpu consumption

  #57 opened 10 months ago by duda1202
  0

• ImportError: cannot import name 'revert_sync_batchnorm' from 'mmcv.cnn.utils'

  #56 opened a year ago by GraceGuoo
  1

• Comparison with SegFormer-B5

  #55 opened a year ago by ArlenCHEN
  1

• AssertionError: train_cfg specified in both outer field and model field

  #54 opened a year ago by wdfcvbmk
  0

• How to start training？

  #53 opened a year ago by sdpscnc
  1

• Inconsistencies in ADE20K model inference output for the same input image.

  #52 opened a year ago by MartaFdez97
  3

• Welcome update to OpenMMLab 2.0

  #51 opened a year ago by vansin
  0

• Config for Pascal VOC

  #30 opened a year ago by jun0kim
  3

• The content of the article is inconsistent with the code

  #46 opened a year ago by Flyingdog-Huang
  1

• Random seeds variance

  #48 opened a year ago by BaiZec
  1

• Is it possible to export SegNeXt as ONNX or tensorrt model to speedup inferencing?

  #25 opened 2 years ago by chenxinfeng4
  2

• when I set the "--gpu-id", default=2, I meet this problem. only "--gpu-id", default=0. I can train the model.

  #47 opened a year ago by zoufeng2416
  0

• the result of SegNeXt training on ADE20k

  #39 opened a year ago by jiawen1929
  1

• Weights for the COCOStuff dataset

  #45 opened a year ago by edwardyehuang
  2

• a question about your paper

  #44 opened a year ago by massica17
  0

• A case that will never happen in /backbones/mscan.py

  #43 opened a year ago by yangtao2019yt
  0

• 有没有不包含mmseg的版本？

  #42 opened a year ago by lindadamama
  0

• Hello, Welcome to the OpenMMLab forum to promote your work

  #40 opened a year ago by vansin
  0

• pad bug, when use pad,iou is error

  #41 opened a year ago by yeyan00
  0

• Encoder-Decoder downsample 1/8x, is too coarse to produce 'seg_logits'

  #35 opened a year ago by chenxinfeng4
  3

• problem about train.py

  #38 opened a year ago by jiawen1929
  8

• Cannot reproduce your results

  #36 opened a year ago by il1um
  3

• about the metrics: 'aAcc' and 'mAcc' in evaluation?

  #37 opened a year ago by Onehundred0906
  1

• ImageNet pretrainning configurations

  #33 opened a year ago by rayleizhu
  3

• Learning Rate and Batch Size

  #32 opened a year ago by FabianSchuetze
  3

• About the training process's error

  #34 opened a year ago by Onehundred0906
  0

• about train？

  #29 opened a year ago by Onehundred0906
  7

• hi ，please help me

  #31 opened a year ago by wudizuixiaosa
  0

• About multi-scale branch

  #28 opened 2 years ago by go-ahead-maker
  1

• Why there is no activation function in attention module?

  #26 opened 2 years ago by RicoSuaveGuapo
  1

• about the inference speed

  #27 opened 2 years ago by Onehundred0906
  1

• Why are the results of Mit's Params not consistent with the results presented in the paper, Segformer?

  #24 opened 2 years ago by HuCaoFighting
  4

• WARNING - The model and loaded state dict do not match exactly

  #23 opened 2 years ago by Yaoxingtian
  1

• Contribute to mmsegmentation

  #18 opened 2 years ago by FabianSchuetze
  5

• how can i do distributed training ?

  #20 opened 2 years ago by Yaoxingtian
  1

• The code implementation does not match the picture in the paper

  #22 opened 2 years ago by pb585
  1

• about the test command(only one gpu)

  #21 opened 2 years ago by Onehundred0906
  2

• about the command of training： ./tools/dist_train.sh /path/to/config 8

  #19 opened 2 years ago by Onehundred0906
  3

• Are params really exact?

  #17 opened 2 years ago by Monkey0806
  2

• how to get mscan_t.pth?

  #16 opened 2 years ago by AnRanbel
  0

• Hello, can you support binary segmentation dataset? 0 or 255(eg: Medical Image Segmentation)

  #15 opened 2 years ago by zwb170010
  4