open-mmlab/mmsegmentation

Issues

• Segformer evaluation metrics

  #3807 opened 2 months ago by MariaMel98
  1

• mmseg config availability

  #3818 opened a month ago by jowaria001
  1

• demo/MMSegmentation_Tutorial.ipynb train error

  #3778 opened 5 hours ago by zhuxuanya
  3

• Error in distributed training/testing

  #3820 opened 5 hours ago by zhuxuanya
  0

• Errors in mask2former caused by class_weight

  #3811 opened 5 hours ago by zhuxuanya
  0

• Is there a conflict between class_weight and seg_ pad_val？

  #3831 opened a day ago by SquirrelEdison
  0

• The problems encountered while training with SAN

  #3796 opened 2 months ago by dongxiaofei12
  2

• Hello, where can I download the pre-trained model of SAN?

  #3830 opened 2 days ago by SaSa-123
  0

• Maintain consistent brightness before and after image inference

  #3829 opened 2 days ago by handsome2456
  0

• How to calculate params and flops of maskformer and mask2former in semantic segmentation?

  #3828 opened 10 days ago by GrayOctopus
  0

• How to integrate SAM architecture into mmseg for finetuning?

  #3827 opened 14 days ago by Hectormxy
  0

• Help!!!【Mask2former】All predictions are colored by full colors in visualization!

  #3826 opened 16 days ago by FreeWillThorn
  1

• Confusing cls_token handling in ViTTransfomer::forward

  #3824 opened 22 days ago by syboomsy
  0

• Question about the pretrained weight configuration for the SAN model

  #3823 opened a month ago by Buchizaoyaowan
  0

• Question about the pretrained weight configuration for the SAN model

  #3822 opened a month ago by qiuyuan666
  0

• Support the implementation of FADC and SCTNet

  #3821 opened a month ago by qiuyuan666
  0

• config settings for SEM images

  #3819 opened a month ago by Mayolov
  0

• When calculating mIoU, the category of 'clutter' was taken into account for mIoU ISPRS potsdam?

  #3816 opened a month ago by kouyuanbo
  0

• 如何对于一个模型进行随机种子的设置呢？

  #3814 opened a month ago by roiyfox
  1

• When I use my own dataset to train the BEIT network, there is no problem with the training process, but the following error is reported during validation, what is the reason?

  #3812 opened a month ago by vGuiGui6
  1

• TypeError: forward() got an unexpected keyword argument 'weight'

  #3815 opened a month ago by llyyccccc
  1

• Export to Torchlib has a mix of Device and Cpu tensors

  #3805 opened a month ago by geoeo
  2

• How to train a custom panoramic segmentation dataset using mmsegmentation

  #3810 opened a month ago by hyx007
  0

• Shape error when running SegFormer

  #3809 opened 2 months ago by sajeedmehrab
  1

• test

  #3804 opened 2 months ago by 23del
  0

• Adding ViT-Adapter as Supported Backbone

  #3802 opened 2 months ago by MatCorr
  0

• TypeError: __init__() missing 1 required positional argument: 'ann_file'

  #3801 opened 2 months ago by Lucky-person
  0

• how can I use "from mmseg.datasets import build_dataset"and"from mmseg.apis import train_segmentor"in the mmseg2.0.0?

  #3800 opened 2 months ago by Reborker
  0

• Add classes issues

  #3784 opened 3 months ago by ParkMinyoung0304
  1

• How to use paseudo lable of train mmseg model or mini change the code to use paseudo label?

  #3799 opened 2 months ago by funny000
  0

• How to use ohem_loss in mmseg==0.29.1?

  #3798 opened 2 months ago by ZYzyio
  0

• How to Add Custom Parameters to loss, predict, and _forward Functions in New Segmentor in mmseg1.2.2

  #3797 opened 2 months ago by jinqin-99
  0

• With the same configuration, the training results of the model vary significantly.

  #3795 opened 2 months ago by guomengqing
  0

• How to fine tune SegFormer on a custom dataset

  #3794 opened 2 months ago by MariaMel98
  0

• pth

  #3793 opened 2 months ago by 23del
  0

• How to use k-fold in the mmseg

  #3789 opened 2 months ago by Vaultaaa000
  0

• How to use K-fold cross-validation in the mmseg 如何在mmseg中使用k-fold方法进行验证

  #3790 opened 2 months ago by Vaultaaa000
  1

• build_backbone

  #3792 opened 2 months ago by dongxiaofei12
  0

• ValueError: need at least one array to concatenate

  #3791 opened 3 months ago by handsome-me
  0

• 预测结果不一致的问题

  #3786 opened 3 months ago by RS-Hong
  1

• problem about reduce_zero_label

  #3787 opened 3 months ago by dongxiaofei12
  0

• How do I build a model using config？

  #3780 opened 3 months ago by dongxiaofei12
  3

• Why not use MMCV==2.2.0?

  #3785 opened 3 months ago by likyoo
  0

• swin-结果疑问

  #3783 opened 3 months ago by MRT0422
  0

• Different decoder channel in original SegFormer paper and MMSegmentation implementation.

  #3782 opened 3 months ago by yhding5618
  0

• get_flops.py can not be used on PointRend

  #3781 opened 3 months ago by Sunny2884
  0

• about loveda dataset config

  #3776 opened 3 months ago by woldier
  0

• Multi-GPU Training in Runner from cfg file

  #3775 opened 3 months ago by RonKat2008
  0

• Resize the pos_embed shape from torch.Size([1, 197, 768]) to torch.Size([1, 1025, 768])

  #3774 opened 3 months ago by whikwon
  0

• How get pred prob of image mask with mmcv2.1.0 and mmseg 1.2.2?

  #3773 opened 3 months ago by funny000
  1