allanj/LayoutLMv3-DocVQA

Issues

• preprocess.extract_spans.py when use MSR OCR

  #11 opened a year ago by sangwu99
  0

• data

  #1 opened 2 years ago by xia-zhenjun-tfs
  13

• Could you provide me with the detailed steps of data processing for docvqa?

  #10 opened a year ago by YULUOYUNZHI
  0

• Could you please give me the processes data of docvqa?

  #9 opened 2 years ago by adda1221
  0

• Could you please share the OCR result using Microsoft OCR API

  #8 opened 2 years ago by qiaoliangzju
  1

• get preprocessed data

  #7 opened 2 years ago by minhoooo1
  0

• pre-processed data

  #6 opened 2 years ago by minhoooo1
  0

• Accuracy == Bad :(

  #3 opened 2 years ago by logan-markewich
  9

• 2-D position embeddings

  #5 opened 2 years ago by StalVars
  2

• How many epochs did it take to achieve the best model for DocVQA?

  #4 opened 2 years ago by StalVars
  0

• RuntimeError: CUDA error: device-side assert triggered

  #2 opened 2 years ago by roburst2
  10