Does anyone test the textsum code to get the perfect result?
Two days ago to see the message xtr33me made " #Textsum# - Incorrect decode results compared to ref file", I now test the situation and this is the same, I use the Chinese corpus, training to running_avg_loss = 0.00001 ,but I decoded the result is wrong,What is the reason?
If anyone can provide any help or direction, it would be greatly appreciated.
check out my latest response here: Textsum - Incorrect decode results compared to ref file
Essentially it all came down to getting more clean data, a little over 1.3million to be exact. Check out the post above should you wish for more of a detailed answer.