chenxin061/pdarts

Issues

• is there a google Colab implementation? is there any chance I can use your Algorithm in a tabular dataset for classification?

  #42 opened 2 months ago by ahmadrasheed
  0

• How to caculate GPU-days(cost)?

  #41 opened 4 years ago by liqier
  0

• The number of filters in each cell

  #40 opened 4 years ago by DaweiXi360
  0

• Reproducing the Results

  #39 opened 4 years ago by kunjing96
  0

• Reproducing the Results

  #38 opened 4 years ago by kunjing96
  0

• CIFAR-100 Performance and Hyperparameters

  #37 opened 4 years ago by jyfliu
  0

• What does 'operation-level Dropout' mean?

  #36 opened 4 years ago by dk-hong
  0

• Question regarding the paper

  #35 opened 4 years ago by yxchng
  1

• parameter stem_multiplier and module stem

  #33 opened 4 years ago by xjhcassy
  6

• why the learning rate and batchsize in "train_imagenet.py" are different from what said in the paper? Which one can reproduce the result？And is the learning rate related to batchsize ?

  #14 opened 4 years ago by AIshenfeng
  1

• How to use this algorithm to my datasets..

  #22 opened 4 years ago by suryakumara
  1

• How many hours on ImageNet of training?

  #9 opened 6 years ago by betterhalfwzm
  11

• PDARTS on medical images classification

  #34 opened 5 years ago by AissamDjahnine
  1

• How many layers we need in the evaluation stage?

  #27 opened 5 years ago by NdaAzr
  11

• lr being passed to train(), but not used

  #32 opened 5 years ago by rtarquini
  1

• semantic segmentation

  #31 opened 5 years ago by NdaAzr
  1

• Add more code comment？

  #6 opened 6 years ago by coolKeen
  3

• Is this a normal case?

  #30 opened 5 years ago by mrk1992
  0

• Question about Dropout Rate in Paper

  #29 opened 5 years ago by volgachen
  4

• confusion about depths of network in different stages

  #28 opened 5 years ago by NdaAzr
  1

• correct RUN command

  #24 opened 5 years ago by xiebangquan
  1

• which genotype is best？

  #21 opened 5 years ago by johsnows
  5

• 关于一阶优化

  #25 opened 5 years ago by hyqyoung
  1

• In the original implementation of DARTS, it is input_search, target_search = next(iter(valid_queue), which slows down the training when using PyTorch 0.4 and above.

  #26 opened 5 years ago by BangguWu
  1

• For other datasets, is it better to use pre-searched network from cifar, or search from itself?

  #13 opened 6 years ago by rtrobin
  13

• Is your method only work on CUDA 10?

  #23 opened 5 years ago by kids0cn
  2

• Multi-gpu support

  #5 opened 6 years ago by R00Kie-Liu
  10

• use train_search , got genotype like DARTS

  #20 opened 5 years ago by ykk648
  2

• if i need train it first and than evalutate?

  #15 opened 5 years ago by johsnows
  4

• Reproducing the Results, and Questions

  #7 opened 6 years ago by drcdr
  22

• In arch search process , why use SGD for operation weight and Adam for arch_params ?

  #18 opened 5 years ago by JarveeLee
  1

• How can resume the train_search.py using weights in log_dir if the code stops due to accidents like power cut ... ?

  #19 opened 5 years ago by JarveeLee
  1

• How can i import the found network architecture ?

  #17 opened 5 years ago by cristyioan2000
  3

• RuntimeError: cuda runtime error (2)

  #16 opened 5 years ago by liu-ca
  0

• How to test?

  #11 opened 6 years ago by zihaozhang9
  4

• Problem while running test.py

  #8 opened 6 years ago by Catosine
  4

• [BUG] test.py

  #12 opened 6 years ago by zihaozhang9
  1

• [Bug]CUDA error: out of memory

  #10 opened 6 years ago by zihaozhang9
  1

• Can i control the flops of the models searched?

  #3 opened 6 years ago by zyc4me
  3

• Missing References

  #4 opened 6 years ago by D-X-Y
  2

• 如果只有ＧＴＸ１０８０ｔｉ（１１Ｇ内存），如何调整参数，让pdarts也可以跑起来训练？

  #2 opened 6 years ago by simonggx
  1