LinWeizheDragon/FLMR

Issues

• Something about paper

  #38 opened 24 days ago by lzhptr
  2

• Something about finetune

  #37 opened a month ago by lzhptr
  4

• No such file or directory

  #36 opened a month ago by XNDIC
  4

• Chinese model and finetuning

  #35 opened a month ago by katie312
  2

• Something about paper

  #34 opened 2 months ago by lzhptr
  5

• Something about evaluate

  #33 opened 2 months ago by lzhptr
  2

• Something about pretraining

  #30 opened 3 months ago by Baeksweety
  11

• difference between FLMR and PreFLMR implementations

  #32 opened 2 months ago by dcaffo98
  2

• Wikipedia Corpus used for OK-VQA task

  #31 opened 2 months ago by ShuaiWang97
  6

• Results compare with InternVit

  #27 opened 4 months ago by lucasjinreal
  1

• does this warning matter? can not find the separator in the input_ids:

  #25 opened 4 months ago by zzk2021
  4

• 结果复现问题

  #26 opened 4 months ago by xiewen354
  2

• Can this model achieve retrieval from text to (image + text)

  #24 opened 5 months ago by Annie1900
  4

• 我从huggingface上下载了LinWeizheDragon / PreFLMR_ViT-B ，但参照示例代码加载模型的时候会报错

  #23 opened 5 months ago by gaoyuanzizizi
  3

• what is pos_item_ids and pos_item_contents mean?

  #21 opened 5 months ago by zzk2021
  1

• problem of finetuning

  #22 opened 5 months ago by haodongze
  1

• 关于index_custom_collection方法中的passage_image_feature的获取方式(How to Obtain Pass_image_feature in the Index_custom_collection Method)

  #20 opened 5 months ago by fengkangjie
  4

• How to do calculate simiarity score between query and gallery item without indexing

  #15 opened 5 months ago by duyc168
  2

• 如何在下游任务中去使用PreFLMR模型？

  #12 opened 5 months ago by fengkangjie
  14

• question about details of finetuning script

  #16 opened 6 months ago by Maxlinn
  9

• 作者你好，这个FLMR和官方版本的Retrieval-Augmented-Visual-Question-Answering区别在哪？FLMR是不是没用blip？

  #19 opened 6 months ago by zzk2021
  2

• confused about the doc_idx，when I split passage_ds and ds, doc_idxis vary large

  #13 opened 6 months ago by zzk2021
  1

• For a custom document, how can I support the input of multiple images?

  #14 opened 6 months ago by yuejunpeng
  1

• Question about image process in customized multi-modal document

  #11 opened 6 months ago by yuejunpeng
  1

• Hello，I want to know the project support windows ？

  #10 opened 7 months ago by zzk2021
  2

• fail to reproduce m2kr infoseek subset results using preflmr model

  #9 opened 7 months ago by Maxlinn
  8

• 请问这个模型可以只传入文本，不传入图片吗？

  #7 opened 7 months ago by jinlong7790
  1

• definition of load_dataset

  #8 opened 8 months ago by duyc168
  2

• load FLMR Retrieval Model Wrong!

  #6 opened 8 months ago by bitccty
  1