CR-Gjx/LeakGAN

Issues

• I can't get the same results as the paper.

  #30 opened 5 years ago by lrt366
  6

• How to create a vocab file for the dataset ? Can you please share the code for it ?

  #35 opened 5 years ago by bharathreddy1997
  0

• A few more instructions in the README

  #8 opened 7 years ago by waynethewizard
  9

• Custom dataset

  #29 opened 5 years ago by emielzyde
  1

• What is vocab_cotra.pkl file and how to generate it?

  #34 opened 5 years ago by bharathreddy1997
  0

• I think the Bootstrapped Rescaled Activation in the code is not useful

  #20 opened 6 years ago by 3015218116
  3

• GPU Usage

  #31 opened 5 years ago by alifurkanbudak
  0

• The code could not run

  #21 opened 6 years ago by maymay1982
  1

• Get “ValueError: No gradients provided for any variable” when running Image_COCO.Main

  #33 opened 5 years ago by CorbinZhou
  0

• Modulo by Zero error when attempting to train on custom dataset

  #32 opened 5 years ago by Dormin92
  0

• What is the latent space in this GAN model?

  #28 opened 6 years ago by ZefanW
  1

• What is the difference between LeakGAN and LM for text generation?

  #22 opened 6 years ago by guotong1988
  0

• Can not understand why use the generated result for the feeding here... Thank you very much...

  #27 opened 6 years ago by guotong1988
  0

• What is the difference between gen_x and the result of rollout?

  #26 opened 6 years ago by guotong1988
  0

• Have you try pretrain the generator by using the next-word-predict loss?

  #23 opened 6 years ago by guotong1988
  0

• Where is the real data inputed?

  #19 opened 6 years ago by mfxss
  1

• Rewards don't change while training

  #17 opened 6 years ago by Seraphli
  7

• Interleaved Training is not used in your code?

  #18 opened 6 years ago by xljhtq
  0

• Number of batches

  #16 opened 6 years ago by akhilvasvani
  0

• ERROR:tensorflow:Couldn't match files for checkpoint ./ckpts/feugan_preD

  #15 opened 7 years ago by Darrenya
  0

• leakgan----generator

  #14 opened 7 years ago by tryinggo
  0

• Where is code for the mentioned "Long Text Generation: EMNLP2017 WMT News" experiment?

  #13 opened 7 years ago by akanimax
  2

• Variable start_token

  #12 opened 7 years ago by Crista23
  1

• Improve LeakGAN by Changing Policy Gradient Structure

  #2 opened 7 years ago by NickShahML
  5

• why there is some 'tail tokens' after the period？

  #11 opened 7 years ago by shaomai00
  2

• In /Image CoCo ，training stop based on what?

  #10 opened 7 years ago by shaomai00
  2

• An issue about your code

  #3 opened 7 years ago by Syrup274
  1

• why the code execution of cosin similarity in the optimization of manger and worker is 1 - cos(dis_feat, goal)?

  #9 opened 7 years ago by qingchenwuhou
  1

• Why CNN instead of LSTM?

  #7 opened 7 years ago by AranKomat
  4

• a little confused to bleu

  #6 opened 7 years ago by hscspring
  2