nlpyang / BertSum

Code for paper Fine-tune BERT for Extractive Summarization
Apache License 2.0
1.47k stars 423 forks source link

Summary files not found #105

Open KarlDavidson opened 4 years ago

KarlDavidson commented 4 years ago

I am presently on the validation step of the process listed in the README. It seems to get most of the way through the process with the temp folder, but I encounter the following error

Exception: Did not find any files matching the pattern cand.(\d+).txt in the system summaries directory Cloned_BERTSUM/temp/tmp7hz2zt07/system.

I can't figure out where this file is supposed to be written in the code?

All my summaries of the validation set are given in a .candidate file, but then when it goes to do things in the temp folder it doesn't know where the summaries are.

Has anyone else had such an issue and fixed it?

tschomacker commented 4 years ago

I ran into the same exception. In my case the error was caused by malformed .story files. The correct format is like this:

text

@highlight

summary

my faulty ones for instance looked this:

text

@highlight summary