Haiyang-W/DSVT

Issues

• full deployment code

  #71 opened 10 months ago by CodeWorldChanged
  8

• NaN problem

  #80 opened 2 months ago by lyhnsn
  2

• Pretrained model

  #79 opened 3 months ago by MSunDYY
  1

• how to train custom dataset

  #56 opened a year ago by evil-master
  2

• Questions about voxels?

  #78 opened 3 months ago by git-xuefu
  2

• Different voxels sizes

  #77 opened 5 months ago by ahmedius2
  1

• nuscenes data visualization fails

  #76 opened 5 months ago by Ultran-cyf
  1

• DSVT-P trainning on Kitti Dataset

  #64 opened a year ago by dinvincible98
  20

• Sparse Shape

  #75 opened 7 months ago by YuRui-Learning
  0

• About Model Params

  #74 opened 8 months ago by ygy123first
  1

• A bug in transfusion_head.py code, in case that pc_range is rectangle

  #73 opened 9 months ago by zzwforever
  2

• Question about multi_frame training in Waymo

  #72 opened 9 months ago by YoushaaMurhij
  4

• DSVT: self-attention key_padding_mask dtype

  #54 opened a year ago by d33dler
  4

• Question TensorRT deployment

  #68 opened a year ago by Anutlet
  1

• 兄弟，你们那个68.4用了tta了吗

  #70 opened a year ago by rubbish001
  4

• dsvt export

  #69 opened a year ago by yumendecc
  1

• Question about the setting of voxel version

  #67 opened a year ago by 3bobo
  2

• Question TensorRT deployment & PCDET

  #60 opened a year ago by d33dler
  5

• Edge computing device deployment issues

  #66 opened a year ago by cyhasuka
  1

• Nuscenes train

  #65 opened a year ago by 123susu
  1

• dsvt_pillar.yaml参数相关问题

  #62 opened a year ago by ZoangX
  3

• DSVT all in tensorRT

  #52 opened a year ago by jingyue202205
  1

• DSVT-TRT deployment dynamic & static shaping

  #61 opened a year ago by d33dler
  1

• KITTI dataset

  #59 opened a year ago by thatnn
  7

• Batch size vs learning rate

  #58 opened a year ago by d33dler
  1

• Dimfit implications

  #57 opened a year ago by d33dler
  1

• Fail to reproduce DSVT-P on nuscenes dataset.

  #55 opened a year ago by synsin0
  19

• DSVT tensorrt model slower than pytorch

  #53 opened a year ago by realwenpeng
  2

• During the model training process, an error occurs in the forward propagation phase. The error message indicates a dimension mismatch between the feature processed by the attention mechanism in dsvt.py and the original feature.

  #51 opened a year ago by luoxiaoliaolan
  6

• Errors included in the process of assigning the Position embedding vector for multi-head self attention

  #50 opened a year ago by junhyung-SPALab
  2

• DynamicPillarVFE

  #49 opened a year ago by sylcito
  1

• param

  #48 opened a year ago by Wang0203
  5

• TensorRT show no improvement in inference speed

  #43 opened a year ago by LinuxCup
  13

• TensorRT Deployment for nuScenes Trained Model

  #46 opened a year ago by yusufy1ld1z
  2

• IndexError in test using nuscenes

  #45 opened a year ago by cjfghddl0206
  5

• Can you provide a configuration file for the KITTI dataset?

  #44 opened a year ago by Rocklei-pu
  1

• Deploy Problem 2

  #42 opened a year ago by sylcito
  3

• DSVT Voxel Performance using 100% waymo dataset

  #41 opened a year ago by ylwhxht
  1

• Timeout when training

  #39 opened a year ago by Wang0203
  5

• Deploy Problem

  #40 opened a year ago by sylcito
  1

• runtime ERROR when compiling ingroup_inds

  #33 opened 2 years ago by tyutlk
  8

• setup error

  #35 opened 2 years ago by Wang0203
  3

• fps of the model

  #38 opened a year ago by PleasantPhoenix
  1

• Metric results for mini-nuscenes

  #37 opened a year ago by Wang0203
  1

• DynPillarVFE in nuscenes setup

  #36 opened 2 years ago by Flamyx
  1

• Loss is nan while training with --fp16(A lots of times)

  #31 opened 2 years ago by ylwhxht
  10

• Detection ability drops significantly at long distances

  #32 opened 2 years ago by zizhengu
  1

• Will the tensorrt version of the code be considered open source?

  #34 opened 2 years ago by christmas-ww
  1

• DSVT Pillar Performance using 100% waymo dataset

  #30 opened 2 years ago by dinvincible98
  3

• Could not reproduce the precision on 20% Waymo

  #29 opened 2 years ago by JingweiZhang12
  5