EMNLP2017_DOC icon indicating copy to clipboard operation
EMNLP2017_DOC copied to clipboard

the experiments setting for 20news

Open xiaoheizai opened this issue 5 years ago • 1 comments

Dear leishu: When I reproduce the code using pytorch, I find that the result of 20news is lower under the 25% setting. I guess it is caused by inconsistent data processing,so I hope to know the details of the data processing of 20news. Thank you !

xiaoheizai avatar Dec 19 '19 09:12 xiaoheizai

(1)use every word inside the document. (2)build vocabulary (pre-trained embedding) based on both training and testing data. Please refer to the code regarding other settings.

Dear leishu: When I reproduce the code using pytorch, I find that the result of 20news is lower under the 25% setting. I guess it is caused by inconsistent data processing,so I hope to know the details of the data processing of 20news. Thank you !

leishu02 avatar Dec 23 '19 10:12 leishu02