valeoai/SLidR

Issues

• mIoU and fwIoU calculation

  #36 opened 4 months ago by barzanisar
  3

• Fusion about the results in README.

  #34 opened 8 months ago by NinjaCXX
  1

• RuntimeError: merge_sort: failed to synchronize: cudaErrorIllegalAddress: an illegal memory access was encountered

  #8 opened a year ago by jaycheney
  13

• None

  #33 opened a year ago by whuhxb
  0

• Object detection experiments

  #32 opened a year ago by TimGor1997
  1

• Setting of the fraction of the training labels

  #30 opened a year ago by gbliao
  2

• Can't pickle local object 'PretrainDataModule.train_dataloader.<locals>.<lambda>'

  #31 opened a year ago by Sean8504
  1

• About training time

  #6 opened 2 years ago by BoPang1996
  5

• Downstream Object Detection Experiments

  #29 opened 2 years ago by youquanl
  2

• Can not reproduce the results in PV-RCNN for object detection. Voxel size different?

  #28 opened 2 years ago by mu-cai
  1

• Random initialization in linear probing

  #26 opened 2 years ago by ruomingzhai
  1

• Object detection experiments

  #15 opened 2 years ago by BoPang1996
  2

• detection on kitti using Minkowski SR-Unet

  #18 opened 2 years ago by Vitarie
  3

• One confusing conv layer in Res16UnetBase

  #27 opened 2 years ago by martin-liao
  0

• Maybe a Bug: Missing Feature Normalize in VoxelNet

  #24 opened 2 years ago by BoPang1996
  3

• Could you please share the visualised code?

  #25 opened 2 years ago by jaycheney
  2

• How to use this project to complete the distillation of target detection method

  #23 opened 2 years ago by VsionQing
  1

• About the the number of superpixels

  #22 opened 2 years ago by jaycheney
  4

• Nuscenes dataset selection for downstream task of semantic segmentation

  #21 opened 2 years ago by nelson-wpwang
  1

• problems of Semantic segmentation

  #19 opened 2 years ago by fire-2019
  5

• problem of the randomness of dectection experiment

  #20 opened 2 years ago by xiahongchi
  0

• The calculation method of mAP in object detection experiment

  #17 opened 2 years ago by xiahongchi
  4

• PPKT pretrain model

  #16 opened 2 years ago by BoPang1996
  0

• Object detection experiments

  #13 opened 2 years ago by jaycheney
  2

• Object detection experiments

  #7 opened 2 years ago by nazMahmoud
  3

• SLIC instead of 2D segmentation

  #14 opened 2 years ago by Stephanie-Shen324
  1

• training loss

  #12 opened 2 years ago by Stephanie-Shen324
  2

• Random initialisation learning rate

  #11 opened 2 years ago by Stephanie-Shen324
  4

• Downstream.py bug: RuntimeError: grad can be implicitly created only for scalar outputs

  #9 opened 2 years ago by Stephanie-Shen324
  2

• I wonder whether there is a bug around line [https://github.com/valeoai/SLidR/blob/main/pretrain/dataloader_nuscenes_spconv.py#L321](https://github.com/valeoai/SLidR/blob/main/pretrain/dataloader_nuscenes_spconv.py#L321 )? (You should not comment this line) Because in your object detection code base, you use 4 features in object detection.

  #5 opened 2 years ago by mu-cai
  3

• help: cannot reproduce segmentation results

  #1 opened 3 years ago by celeste-1
  3

• The train loss fluactuates at 7 but does not decrease

  #4 opened 2 years ago by ruomingzhai
  5

• Training is stuck for pretraining

  #3 opened 2 years ago by mu-cai
  8

• Question about reproducing experiment

  #2 opened 3 years ago by opjang5
  1