yongliu20/QDMN

Issues

• About Score class

  #19 opened 6 months ago by pengzz-ops
  0

• About Score class

  #18 opened 6 months ago by pengzz-ops
  0

• The code has been reproduced, but the results are inconsistent with the provided model.

  #17 opened 6 months ago by qfxb
  0

• Adding QAM only in STM's inference stage

  #15 opened 2 years ago by csustYyh
  1

• 为什么eval代码要用到label?

  #16 opened 2 years ago by yanghgai
  0

• youtube2018

  #14 opened 2 years ago by karchen-geng
  2

• 您好,当我将QAM加入到STCN当中时,且输入到Score模块的特征是fuse后的,但是训练得到的score值到后面都是0了(Score模块里面全连接层的激活函数一个是relu,一个是sigmoid)我是1GPU训练,batch size为16,学习率为1e-5

  #7 opened 2 years ago by lzl2040
  7

• train.py启动问题

  #9 opened 2 years ago by karchen-geng
  5

• May ask the result when you use the global average pooling layers in the code of QAM module?

  #13 opened 2 years ago by zhouweii234
  1

• Do you train the whole network when you apply QAM in STCN?

  #12 opened 2 years ago by zhouweii234
  1

• Questions about pre-training

  #11 opened 2 years ago by karchen-geng
  19

• 关于预训练的问题

  #10 opened 2 years ago by karchen-geng
  3

• checkpoint file

  #8 opened 2 years ago by nankepan
  1

• Is the input dimension of Score 256 when QAM is used in STCN?

  #1 opened 2 years ago by lzl2040
  3

• 请问预训练阶段是不需要调整学习率的吗？代码中预训练时的steps比它的迭代次数还要大

  #6 opened 2 years ago by lzl2040
  6

• 您好，注意到您的代码变了，请问是之前的代码有问题吗？

  #5 opened 2 years ago by lzl2040
  2

• 您好，请问论文中为什么不使用L2 distance而是dot product呢，是L2 distance效果比dot product差吗

  #4 opened 2 years ago by lzl2040
  2

• When the QAM is used in STCN,is the iterations the same as STCN?

  #3 opened 2 years ago by lzl2040
  1

• I use QAM for training, the number of iterations is 75K, but the loss only drops to about 0.5, is this normal?And how much does the final loss value of the original code drop?

  #2 opened 2 years ago by lzl2040
  12