sacmehta/ESPNetv2

Issues

• Can this model be used for regression tasks?

  #39 opened 3 years ago by sunshinewhy
  0

• The tool to calculate FLOPs

  #38 opened 3 years ago by huanghongxing-jim
  0

• Have you used ImageNet pretrained weight to train epsnetv2 for segmentation?

  #35 opened 5 years ago by BaobeiChenchen
  1

• Espnet .pth with another scale factor

  #32 opened 5 years ago by marcusvlc
  1

• Can anyone upload a pretrained model?

  #36 opened 5 years ago by dragen1860
  1

• compared with ShuffleNet V2

  #29 opened 5 years ago by cs-heibao
  3

• Network parameters of segmentation model in the paper

  #34 opened 5 years ago by Jason93K
  5

• You have following values as class labels: [ 0 1 2 4 5 6 7 8 10 11 13 255] Some problem with labels. Please check image file: ./city/gtFine/train/cologne/cologne_000000_000019_gtFine_labelTrainIds.png Exiting!!

  #33 opened 5 years ago by jwnirvana
  2

• Question about predict on espnetv2

  #31 opened 5 years ago by marcusvlc
  3

• train/eval for another dataset

  #30 opened 5 years ago by marcusvlc
  1

• Using sum of 2 losses

  #15 opened 5 years ago by jongmokim7
  2

• A strange problem encountered during training

  #25 opened 5 years ago by monk42
  15

• Turning ESPNet into c++ cudastream problem

  #27 opened 5 years ago by engineer1109
  2

• About Power consumption

  #26 opened 6 years ago by stigma0617
  9

• train.txt create

  #28 opened 6 years ago by yuliang1995
  2

• Downsampling

  #20 opened 6 years ago by olegarch
  1

• A strange problem encountered during training

  #24 opened 6 years ago by monk42
  0

• a strange question

  #23 opened 6 years ago by monk42
  0

• a strange question

  #22 opened 6 years ago by monk42
  0

• Training issues (from pretrained weights)

  #21 opened 6 years ago by ToniRV
  1

• Is it possible to get probability(likelihood) for foreground class label?

  #18 opened 6 years ago by c13proto
  2

• How to train own dataset?

  #13 opened 6 years ago by pentaris
  3

• About cyclic learning rate

  #19 opened 6 years ago by lizhengwei1992
  1

• numpy is not imported in Dataset.py

  #17 opened 6 years ago by c13proto
  2

• Performance Issues on NVIDIA GTX1080

  #16 opened 6 years ago by nithishc829
  10

• get 8% miou when train ade datasets with 7classes

  #14 opened 6 years ago by TianDe11
  2

• Classes Number maybe not reasonable

  #11 opened 6 years ago by TianDe11
  12

• Question about the comparison between ESPNet and ESPNetv2 on your paper.

  #12 opened 6 years ago by msson
  2

• What does 's' mean?

  #10 opened 6 years ago by wrpha-group
  6

• I cannot get your accuracy after following your steps to train on the cisyscapes datase.

  #9 opened 6 years ago by gao-ye
  16

• RuntimeError: cuda runtime error (59) : device-side assert triggered at /pytorch/aten/src/THC/generic/THCTensorCopy.cpp:70

  #4 opened 6 years ago by monk42
  12

• Is there a problem here ?

  #8 opened 6 years ago by monk42
  3

• hi，when i train segmentation with two classes

  #7 opened 6 years ago by zhangyunming
  9

• RuntimeError: invalid argument 0: Sizes of tensors must match except in dimension 1. Got 136 and 135 in dimension 2 at /opt/conda/conda-bld/pytorch_1532581333611/work/aten/src/THC/generic/THCTensorMath.cu:87`

  #6 opened 6 years ago by tokyokuma
  4

• What is the actual latency on ARM or CPU?

  #5 opened 6 years ago by zeyu-liu
  2

• When run segmentation codes, I got a error!

  #3 opened 6 years ago by monk42
  0

• Cityscape Dataset Error

  #2 opened 6 years ago by monk42
  6

• MNasNet

  #1 opened 6 years ago by pool1892
  1