YudeWang/deeplabv3plus-pytorch

Issues

• Pretrained Xception

  #57 opened 3 years ago by RYY0722
  6

• multi-scale test severely damage the performance

  #44 opened 5 years ago by XushengLee
  2

• after test，only background is defected

  #40 opened 5 years ago by pengyaru
  2

• How to inference with any images?

  #56 opened 4 years ago by snowmint
  0

• Performance is lower than the results you report.

  #48 opened 5 years ago by zhouyuan888888
  9

• Why not change the dilation of this to 2 here ?

  #54 opened 4 years ago by CoinCheung
  0

• 模型在cityscapes数据集上的性能问题

  #53 opened 4 years ago by pingjun18-li
  1

• voc训练数据只有1464张图的时候训练精度是多少呢

  #50 opened 4 years ago by niurouli
  1

• voc test.py mIOU value is error

  #49 opened 4 years ago by cwwng
  11

• how i use my own dataset？

  #30 opened 5 years ago by memeda2232
  2

• imagenet xception pretrained weights

  #52 opened 5 years ago by sbelharbi
  1

• Training and Testing with ADE20K Dataset

  #51 opened 5 years ago by binging512
  2

• ValueError: Expected more than 1 value per channel when training, got input size torch.Size([1, 256, 1, 1])

  #34 opened 5 years ago by zdaiot
  1

• The performance paper reported is 89% mIoU on PASCAL VOC 2012

  #46 opened 5 years ago by RunpeiDong
  4

• The pretrained model of xception

  #47 opened 5 years ago by zhouyuan888888
  0

• How can I get mIoU=79.916%?

  #43 opened 5 years ago by He-Chao
  15

• training error

  #45 opened 5 years ago by JackYang1008
  1

• Error with 'sync_batchnorm' when batchsize=1

  #39 opened 5 years ago by Flames60
  0

• follow your training cfg, deeplabv3+res101 can only achieve 76.777% miou

  #31 opened 5 years ago by whitesockcat
  14

• 关于ASPP的结构，有疑问

  #41 opened 5 years ago by zhangbanxian123
  1

• VOC数据集里的白色mask部分被算作了背景

  #42 opened 5 years ago by XushengLee
  7

• 你好，请问有cityscapes的评估代码吗？

  #38 opened 5 years ago by SecondLi
  1

• hello，请问img为什么没有减mean除以std？

  #37 opened 5 years ago by opee007
  5

• The pretrained model on cityscapes

  #36 opened 5 years ago by MandyMo
  0

• cannot use xception backbone with pretrained checkpoint

  #35 opened 5 years ago by jt78kim
  11

• torch version erro

  #32 opened 5 years ago by allendred
  3

• single GPU inference error

  #33 opened 5 years ago by hellozgm
  4

• the difference between voc and vocfinetuning?

  #28 opened 5 years ago by NJU-yasuo
  13

• xception pretrained models

  #18 opened 6 years ago by spytensor
  6

• question on tensor allocation

  #29 opened 5 years ago by tsing90
  4

• Two question about transform on dataset

  #26 opened 5 years ago by Balabala-Hong
  3

• Cityscapes classes?

  #27 opened 6 years ago by DRosemei
  1

• xception pretrain on coco

  #6 opened 6 years ago by 666zz666
  4

• Learning rate for BatchNorm layers

  #23 opened 6 years ago by sjbz1926
  1

• xception training time

  #21 opened 6 years ago by Colorfu1
  1

• support pytorh-1.0?

  #25 opened 6 years ago by zimenglan-sysu-512
  1

• resnet pretrained model

  #24 opened 6 years ago by XingLiuJia
  0

• deeplabv3+ VOC

  #22 opened 6 years ago by Wangyf46
  1

• VOC augmented segmentation dataset

  #20 opened 6 years ago by Wangyf46
  3

• About atrous/Multi-Grid

  #19 opened 6 years ago by Spritea
  2

• How to use my own dataset

  #16 opened 6 years ago by MissAimee
  2

• Performence on cityscape

  #14 opened 6 years ago by neineimeow
  8

• Single GPU?

  #15 opened 6 years ago by Yuliang-Zou
  1

• pretrained model

  #13 opened 6 years ago by zmghub
  1

• problem of running code

  #12 opened 6 years ago by zwtburger
  1

• Some questions about dataset

  #11 opened 6 years ago by neineimeow
  1

• How much time do 46 epochs cost for pascal voc 2012

  #10 opened 6 years ago by ykp1992
  8

• net.module in train.py

  #9 opened 6 years ago by ykp1992
  4

• different TRAIN_BN_MOM for COCO and VOC

  #8 opened 6 years ago by ChiWeiHsiao
  2

• Core Dump error

  #7 opened 6 years ago by neineimeow
  6