LantaoYu/SeqGAN issues and pull requests

#71 - Questions about the recurring results of the code

Issue - State: open - Opened by giraffeCjl over 3 years ago - 1 comment

#70 - How to resume training in Colab?

Issue - State: open - Opened by syedasara-angelium almost 4 years ago

#69 - How can i use my own training data ?

Issue - State: open - Opened by dieey over 4 years ago - 2 comments

#68 - dataset

Issue - State: open - Opened by saharjandaghy almost 5 years ago - 1 comment

#67 - NLG

Issue - State: open - Opened by nlpofwhat about 5 years ago

#66 - About the accuracy of discriminator during training?

Issue - State: open - Opened by xieexiaotuzi about 5 years ago

#65 - data format

Issue - State: open - Opened by ExpressGit about 5 years ago

#64 - How should I understand the RL loss function

Issue - State: closed - Opened by yanghoonkim over 5 years ago

#63 - About generator in adversarial training

Issue - State: open - Opened by xieexiaotuzi over 5 years ago - 2 comments

#62 - gradient decent implementation

Issue - State: open - Opened by o20021106 over 5 years ago

#61 - If the positive sample and negative sample of each training are corresponding, will it affect the training result

Issue - State: open - Opened by yeshenpy over 5 years ago

#60 - About oracle model

Issue - State: open - Opened by jaingmengmeng over 5 years ago

#59 - About dataset.

Issue - State: closed - Opened by jaingmengmeng over 5 years ago - 1 comment

#58 - Nothing. Ignore it.

Issue - State: open - Opened by Angus07 over 5 years ago

#57 - What is the difference between SeqGAN and LM for text generation?

Issue - State: open - Opened by guotong1988 over 5 years ago - 1 comment

#56 - Updated sequence_gan to be Python 3.6 compatible

Pull Request - State: open - Opened by Deniall over 5 years ago - 3 comments

#55 - 关于采样的问题

Issue - State: open - Opened by J-zin almost 6 years ago

#54 - Training on custom dataset

Issue - State: closed - Opened by WorldHellow almost 6 years ago - 1 comment

#53 - the input data?

Issue - State: open - Opened by YiLing28 almost 6 years ago

#52 - 模型问题

Issue - State: closed - Opened by TyrionChou about 6 years ago

#51 - About the sequences to process

Issue - State: open - Opened by sagechen about 6 years ago

#50 - Keras implementation of SeqGAN

Issue - State: open - Opened by rahulvigneswaran about 6 years ago

#49 - how can i run this code with python3.5? i dont know how install tensorflow-gpu in python2.7

Issue - State: open - Opened by bikuku about 6 years ago - 2 comments

#48 - Issue regarding new review generation

Issue - State: open - Opened by akhileshkumargangwar about 6 years ago

#47 - Question about the corpus

Issue - State: open - Opened by wyxcc over 6 years ago - 2 comments

#46 - Confusion about generator training in adversarial train

Issue - State: closed - Opened by xiyan524 over 6 years ago

#45 - About baseline of reward function

Issue - State: open - Opened by jackyyeh5111 over 6 years ago - 1 comment

#45 - About baseline of reward function

Issue - State: open - Opened by jackyyeh5111 over 6 years ago - 1 comment

#44 - Three questions about this model ~

Issue - State: open - Opened by redinton over 6 years ago - 1 comment

#43 - Why do you define a create_recurrent_unit function yourself？use lstm?

Issue - State: open - Opened by LLLLLLoki over 6 years ago - 1 comment

#42 - Modify get_reward to suit various length sequence

Pull Request - State: closed - Opened by TobiasLee over 6 years ago

#41 - what the 'start state s0' exactly is?

Issue - State: open - Opened by shaomai00 over 6 years ago - 1 comment

#40 - Is the loss of the discriminator in GAN the same as that in pretraining?

Issue - State: closed - Opened by eduOS over 6 years ago - 1 comment

#39 - Monte-Carlo roll-out

Issue - State: closed - Opened by WangWenshan over 6 years ago - 4 comments

#38 - Vocab Dictionary

Issue - State: closed - Opened by EternalFeather almost 7 years ago - 1 comment

#37 - 输入数据在哪?

Issue - State: open - Opened by Perfeee almost 7 years ago - 2 comments

#36 - Could you please provide source code about music generation?

Issue - State: open - Opened by Albert-Ye almost 7 years ago

#35 - Once the training is done, how can we reuse the model?

Issue - State: open - Opened by sweetkurapika almost 7 years ago - 1 comment

#34 - About pre-training generator

Issue - State: open - Opened by playma about 7 years ago - 1 comment

#33 - Understanding of reward and loss function

Issue - State: closed - Opened by tocab about 7 years ago - 7 comments

#32 - Why do you use a separate rollout instead of directly using the Generator?

Issue - State: open - Opened by kunrenzhilu about 7 years ago - 1 comment

#31 - How the gradients pass down if there's a tf.multinomial() sampling process?

Issue - State: closed - Opened by kunrenzhilu about 7 years ago - 1 comment

#30 - where does the randomness of the generator come from?

Issue - State: closed - Opened by kunrenzhilu about 7 years ago - 1 comment

#29 - why should update rollout policy in this way?

Issue - State: closed - Opened by vanpersie32 about 7 years ago - 8 comments

#28 - question about update_params in roll out policy

Issue - State: open - Opened by vanpersie32 about 7 years ago

#27 - Did you track discriminator accuracy and loss?

Issue - State: open - Opened by benstaf about 7 years ago - 1 comment

#26 - problems happened

Issue - State: open - Opened by daveanyan about 7 years ago

#25 - How to run the code in the python3.5.2

Issue - State: closed - Opened by daveanyan about 7 years ago

#24 - I still don't know exactly what (N Time Monte Carlo sampling) is..

Issue - State: closed - Opened by guotong1988 about 7 years ago - 1 comment

#23 - What is the function of rollout.update_params()?

Issue - State: closed - Opened by guotong1988 about 7 years ago - 2 comments

#22 - Support python3.5.2?

Issue - State: closed - Opened by wangtao1321 about 7 years ago - 1 comment

#21 - strange behavior of reward signal

Issue - State: open - Opened by lederg about 7 years ago - 2 comments

#21 - strange behavior of reward signal

Issue - State: open - Opened by lederg about 7 years ago - 2 comments

#20 - about the loss of generator

Issue - State: closed - Opened by wjb123 about 7 years ago - 2 comments

#19 - Question about Rollout

Issue - State: closed - Opened by nathan-whitaker over 7 years ago - 3 comments

#18 - Training strategy?

Issue - State: closed - Opened by zhangzibin over 7 years ago - 1 comment

#17 - Int to Word Mapping needed

Issue - State: closed - Opened by caffeine96 over 7 years ago - 2 comments

#16 - View text data in result

Issue - State: closed - Opened by rahulptel over 7 years ago - 1 comment

#15 - target_params.pkl open using pyhotn3 pickle failed.

Issue - State: closed - Opened by lucasjinreal over 7 years ago - 6 comments

#14 - Sequential data as vectors?

Issue - State: closed - Opened by daking2014 over 7 years ago - 1 comment

#13 - target_params.pkl encode

Issue - State: closed - Opened by jianqiaoc over 7 years ago - 1 comment

#12 - evaluation issues

Issue - State: closed - Opened by xiaopyyy over 7 years ago - 6 comments

#11 - Fixed typo that breaks the Seq GAN

Pull Request - State: closed - Opened by andreasveit over 7 years ago

#10 - Loss of the generator

Issue - State: closed - Opened by ghost over 7 years ago

#9 - Ui should be a matrix from hidden state dim to hidden state dim

Pull Request - State: closed - Opened by jozef-mokry over 7 years ago

#8 - nit, global_variable_initializer is a function

Pull Request - State: closed - Opened by sygi over 7 years ago

#7 - global_variables_initializer error - tensor flow 12.1

Issue - State: closed - Opened by GenTxt over 7 years ago - 4 comments

#6 - How to get the parameters of a different target_lstm?

Issue - State: closed - Opened by xiaopyyy over 7 years ago - 7 comments

#5 - Updated to new TensorFlow, added GitIgnore

Pull Request - State: closed - Opened by Nopey over 7 years ago

#4 - AttributeError: 'module' object has no attribute 'While'

Issue - State: closed - Opened by SeekPoint almost 8 years ago - 1 comment

#3 - How to train and generate from our own Data?

Issue - State: closed - Opened by ghost almost 8 years ago - 2 comments

#2 - the control_flow_ops.While has been changed to tf.while_loop

Issue - State: closed - Opened by ruotianluo almost 8 years ago - 3 comments

#1 - Please Cite

Issue - State: closed - Opened by ofirnachum about 8 years ago - 2 comments

GitHub / LantaoYu/SeqGAN issues and pull requests