Closed darrayes closed 6 years ago
1) My dataset is the same CNN but the way labels are found is different. hence the dataset is essentially different.
What labels are you talking about? Could you explain, I am not able to help with the provided information.
2) I don't want to limit to three sentences in full length evalaution.
data_utils calls a funtion called: predict_toprankedthree from model_utils. You have to replace this function or change values in this function.
I mean to say the way you label a sentence as being eligible for summary or not (0 or 1) is different for me.
We treat it as a sequence labeling task using binary labels.
Hi, I want to compare the results with my own system and hence I need help in two issues:
Can you help me with these two issues and point to relevant sections of the code.