eecrazy / ConstructingNEEG_IJCAI_2018

The data and code for our IJCAI-ECAI 2018 paper: Constructing Narrative Event Evolutionary Graph for Script Event Prediction

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Interpreting Scores

Bharathi-A-7 opened this issue · comments

When I try to print out the scores, coming from the 'compute_scores()' function, this is what I get :

        [-0.8766, -0.8788, -0.8806, -0.8800, -0.8850],
        [-1.0144, -1.0146, -0.9776, -0.9630, -1.0120],
        ...,
        [-0.9916, -0.9794, -0.9871, -1.0362, -0.9936],
        [-0.8638, -0.8637, -0.9221, -0.8754, -0.8722],
        [-0.9519, -0.9568, -0.9875, -0.9767, -0.9815]], grad_fn=<NegBackward>)

If my understanding is correct, the rows represent each example in the given batch and the columns represent the relatedness scores for each event in the candidates list(There are 5 events), right ?

I'm trying to use your code for a different application, so just wanted to get this clear.

Please let me know if this is correct.

Thanks

commented

yes, you are right