DualRL
DualRL copied to clipboard
luofuli
A Dual Reinforcement Learning Framework for Unsupervised Text Style Transfer (IJCAI 2019)
With those .tsf files and .0 .1 files nice and neat, running the bleu perl program called an error: CalledProcessError: Command 'perl C:\Users\myname\DualRL\utils\script\multi-bleu.perl C:\Users\myname\DualRL/references/yelp/reference0.1 C:\Users\myname\DualRL/references/yelp/reference1.1 C:\Users\myname\DualRL/references/yelp/reference2.1 C:\Users\myname\DualRL/references/yelp/reference3.1 < C:/Users/myname/DualRL/tmp/output/yelp_final/0_test.1.tsf' returned...
hi fuli, I face with "ZeroDivisionError: division by zero" when calculate H_score. Is it caused by warm-up? Do you have some idea?
I am running this model in a Google Colab notebook. However, after following the steps in the ReadMe file to train the model, I'm not sure how I can apply...
Hi, nice repo and I have a little question when reading your code. Why did you take greedy decode results as baseline and use it to normalize random decode results?...
Hi, Fuli. Thanks for your great work. I'm very curious about the generating outputs, how do you get those outputs of the baseline models? Did you try to reproduce the...
Hi Fuli, I have trained the NMT model using OpenNMT=1.15, now trying to restore checkpoints (trained using OpenNMT) in DualRL setting. But facing issue while restoring, **Error: Key NMT0-1/NMT0-1/decoder/dense/bias not...
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...
Hi, thanks for your great work! We were just getting started with the code and were wondering how exactly you generated the templates for the GYAFC dataset. We have all...
Thank you for this great work! It seems it's not straightforward to apply the template-based method to the informal-formal dataset since there're no clear attribute markers as those in the...
