-
Why is it that wassertein loss was not used in adversarial training?
-
I'd like to use a decoder-only Transformer with fast coding ability to use it as a RL agent in NLP like SeqGAN. transformer_moe.py has decoder-only option, but it seems to require onerous modification…
-
Hi, zhen yang, I am confused about the ```g_loss``` in function ```gan_output```(662 in model.py) after reading through your paper and Yu's SeqGAN paper, is this line https://github.com/ZhenYangIACAS/…
-
Hi,
First of all, thanks for sharing your code! I’m impressive of your solid work. However, I found some issues when I run your code under different hyper-parameters.
Main issues:
1. Synthet…
-
**System information**
- Have I written custom code (as opposed to using a stock example script provided in TensorFlow): Yes
- OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Arch Linux
- …
ghost updated
5 years ago
-
Hi,I am so happy to find this project.
But I am confused about the dataset.
May I ask where the dataset come from? And is it a public dataset?
I find no other way to contact you, so I can only use …
-
Please make sure that the boxes below are checked before you submit your issue. If your issue is an implementation question, please ask your question in the [Keras-RL Google group](https://groups.goog…
-
There is no example code for gan training with seq2seq, I know it is somewhat like a combination of seq2seq_attn and seqgan, but still not sure about how to do it specifically. Can someone show me som…
-
-
Experiment with different regularizers for training of existing and future sequence models. This should improve the overall quality of generated sequences, including those generated in real-time where…