Question about Baseline Codes

luofuli / DualRL

A Dual Reinforcement Learning Framework for Unsupervised Text Style Transfer (IJCAI 2019)

https://export.arxiv.org/pdf/1905.10060

MIT License

264 stars 44 forks source link

Question about Baseline Codes #17

Open HZQ950419 opened 4 years ago

HZQ950419 commented 4 years ago

Hi Luofuli, This is a very nice work in text style transfer and I have a question. Are the results of baselines from the code released by original authors or the code reimplemented by you?

luofuli commented 4 years ago

Most of the baselines are from https://github.com/lijuncen/Sentiment-and-Style-Transfer/tree/master/evaluation/outputs/yelp. And I implemented BackTrans and Unpaired using the corresponding open-source code on github. UnsuperMT baseline is implemented by the author of the original paper ( Zhang sent me the generated results by email).

HZQ950419 commented 4 years ago

Thanks for your reply!

zhunipingan commented 4 years ago

学姐好请问您知道这里的output分别是对应于哪些模型的结果吗，您采用的对比方式是把这些输出通过自己的评价方式test得出评价结果吗. 期待指点与回复！谢谢~