lverwimp

Hi, You can use the 'rescore' option for this: add 'rescore' to your configuration file, and as value you can give it the path to a file containing the sentences...

[ https://github.com/lverwimp/tf-lm/blob/master/config/en-ptb_word_discourse_rescore.config ] There is an example on Github: en-ptb_word_discourse_rescore.config You need to change the value of 'True' after rescore to a filename, and than that file will be rescored...

Hi Ken, Could you provide me the config file that you used, as well as the size of the dataset that you are testing on (number of words)? Best, Lyan...

I tried to replicate this with your config file and test set, but in my test the script stopped like it should. The 2 most likely causes are: - in...

What do you mean with +1 at the very end? Can you show me the output?

I fixed an error in the code: there was some confusion between self.cache and self.use_cache. Can you try again with the new code and let me know if it works...

Ok, the issue was not run_epoch.py but in lm.py. I added use_cache to the class `lm'. Can you check whether it works now?

I will probably add tf1.5/tf1.6 support in the future, not sure about Python 3 but I'll think about it. Thanks for the suggestion! Van: "Sanjay Krishna Gouda" Aan: "lverwimp/tf-lm" Cc:...