cshizhe/VLN-HAMT

Issues

• Memory leak during newEpisode in data pre-processing

  #16 opened 8 months ago by GengzeZhou
  0

• The issues about the performance of two-Stage pretraining model

  #15 opened 9 months ago by imzhangsheng
  2

• About the second stage is training ViT in an end-to-end manner.

  #14 opened 9 months ago by yhl2018
  0

• About the pre-training problems

  #13 opened 9 months ago by imzhangsheng
  1

• About the e2e trained ViT model

  #5 opened a year ago by alloldman
  1

• Instructions to obtain 'panoimages.lmdb'

  #8 opened a year ago by zdou0830
  1

• Suggest to loosen the dependency on networkx

  #12 opened 2 years ago by Agnes-U
  0

• Multi-node training

  #11 opened 2 years ago by qizhust
  1

• provided features differs from extracted features

  #9 opened 2 years ago by wz0919
  2

• Could you sepecify the models that can reproduce the reported results?

  #6 opened 2 years ago by Xin-Ye-1
  1

• Question about the version of CLIP used in RxR.

  #7 opened 2 years ago by MarSaKi
  1

• Matterport3D simulators

  #4 opened 2 years ago by diaoyudiaochan
  0

• Questions about ViT end2end training

  #3 opened 2 years ago by jialuli-luka
  4

• Questions about pre-trained model

  #2 opened 2 years ago by jialuli-luka
  1

• Could you please share the running scripts for IL+RL training from scratch?

  #1 opened 2 years ago by Jackie-Chou
  3