ChenRocks/UNITER

Issues

• txt_db preprocessing code for VQA

  #67 opened 4 years ago by tejas-gokhale
  4

• lmdb.Error: /userhome/UNITER/UNITER/img/coco_train2014/all: No such file or directory

  #80 opened 3 years ago by wangxiao5791509
  1

• Is the training data(nlvr2_train.tar) properly compressed?

  #95 opened 2 years ago by cokemhlee
  0

• which is not yet supported in this version of the container

  #90 opened 3 years ago by liupeng0606
  1

• pretrained model

  #94 opened 2 years ago by FightingFighting
  0

• About fp16 and location features.

  #93 opened 2 years ago by EstherBear
  0

• prepare img / txt db code for CC / SBU

  #64 opened 4 years ago by HarveyGuo960817
  3

• AttributeError: 'Namespace' object has no attribute 'train_datasets'

  #92 opened 3 years ago by ForawardStar
  2

• Could not find the decoder in codes

  #91 opened 3 years ago by longkukuhi
  1

• AttributeError: 'DetectFeatLmdb' object has no attribute 'env'

  #70 opened 4 years ago by FightingFighting
  5

• How to judge the convergence of the pre-training model？

  #89 opened 3 years ago by PamelaDDD
  0

• What is the gather index?

  #86 opened 3 years ago by abhidipbhattacharyya
  1

• broadcast_tensors get stuck

  #50 opened 4 years ago by tjulyz
  2

• how to generate image feature from ground true?

  #87 opened 3 years ago by FightingFighting
  1

• Has anyone tried not to use a container environment?

  #88 opened 3 years ago by PamelaDDD
  1

• if you are going to provid the code about extracting feature from a image?

  #72 opened 4 years ago by FightingFighting
  7

• is it hard to reimplement UNITER for reffering expression task?

  #83 opened 3 years ago by SkylerZheng
  1

• SBU train/val splits

  #85 opened 3 years ago by erobic
  0

• How is re_coco_gt calculated? Difference between re_coco_det and re_coco_gt

  #84 opened 3 years ago by statechular11
  0

• is it possible not to use docker? It's really hard to debug with docker on remote server. Could you provide the requirements to reproduce the experiment without using docker?

  #81 opened 3 years ago by SkylerZheng
  0

• Reproduce error: Non-existant physical address

  #82 opened 3 years ago by jieWANGforwork
  0

• Preprocessing code for SNLI-VE

  #52 opened 4 years ago by Worm4047
  2

• ImportError: cannot import name 'MlmDatasetForVCR' from 'data' (/media/wangxiao/Samsung_P2/UNITER/data/__init__.py)

  #79 opened 3 years ago by wangxiao5791509
  0

• missing image dbs of vg in the download_indomain.sh

  #73 opened 4 years ago by RenShuhuai-Andy
  0

• Is it possible to extract features using 11GB GPU?

  #78 opened 4 years ago
  1

• What format is used when extracting features, RGB or BGR?

  #77 opened 4 years ago
  1

• Why we need to add size_mul here?

  #53 opened 4 years ago
  3

• feature extraction with docker: RuntimeWarning: invalid value encountered in greater_equal

  #76 opened 4 years ago by suzyahyah
  0

• Is it possible to release model checkpoints pretrained on out-of-domain data (CC, SBU) only?

  #75 opened 4 years ago by Mallory24
  0

• Pretraining from scratch on small data

  #74 opened 4 years ago by yonatanbitton
  0

• Can't run preprocessing

  #66 opened 4 years ago by steremma
  1

• img db gt features

  #71 opened 4 years ago by fromSun2Moon
  0

• Hyperparameters for image feature extraction

  #69 opened 4 years ago by senzeyu
  0

• Image Text Retrieval Loss Discussion

  #68 opened 4 years ago by jipson7
  0

• Location features

  #65 opened 4 years ago by FuyuWang
  0

• error in download_nlvr2.sh (while extracting tar files)

  #63 opened 4 years ago by tejas-gokhale
  1

• Could you share the finetuned weights for image-text retrieval?

  #62 opened 4 years ago by hwanheelee1993
  0

• Is there code to reproduce the "Referring Expression Comprehension" results?

  #61 opened 4 years ago by prashan
  0

• Any faster feature extraction tools (better than the one in bottom-up-detection and lxmert)?

  #60 opened 4 years ago by yezhengli-Mr9
  2

• *.mdb file's consulting

  #59 opened 4 years ago by luoying050601
  0

• pre-trained model

  #57 opened 4 years ago by yixuan-qiao
  1

• When I set num_worker=8, the following error will occurs

  #58 opened 4 years ago by youngfly11
  0

• the vocab file in uniter

  #56 opened 4 years ago by wjy3326
  1

• how to finetune in my own dataset?

  #55 opened 4 years ago by wjy3326
  3

• Uniter's IMG_ LABEL_ DIM

  #54 opened 4 years ago by wjy3326
  1

• RefCOCO training / evaluation details

  #51 opened 4 years ago by j-min
  2

• VCR Second Stage Pretraining

  #49 opened 4 years ago by JaredFern
  2

• Vocabulary and single image-question pair prediction

  #46 opened 4 years ago by foxm79
  2

• Could you provide visualization code?

  #48 opened 4 years ago by jaeyun95
  0

• Could your provide the initial model 'uniter-base-init.pt' for uniter pretraining?

  #47 opened 4 years ago by tjulyz
  0