Skip to content
This repository has been archived by the owner on Oct 31, 2023. It is now read-only.

Code in datasets.py #2

Open
ling-cai opened this issue Sep 29, 2020 · 5 comments
Open

Code in datasets.py #2

ling-cai opened this issue Sep 29, 2020 · 5 comments

Comments

@ling-cai
Copy link

in datasets.py 'while id_timeline < len(self.events) and self.events[id_timeline][0] <= eval_events[id_event][3]:' should it be 'eval_events[id_event][0]' instead of 'eval_events[id_event][3]'?

@ling-cai
Copy link
Author

ling-cai commented Sep 29, 2020

Yes. i guess this is to filter out some positive predictions to get filtered evaluation metrics. However,
self.events[id_timeline][0] --> this is the date (start/end) associated with the event id 'id_timeline'.
but I checked on the eval_events[id_event][3],which is the object id. So i am confused. Not sure whether this is a bug or I misunderstood something.

@timlacroix
Copy link
Contributor

Actually you're right.
let me fix this and check what happens

@ling-cai
Copy link
Author

yes. i agree that it would not make a big difference to the results. Just wanted to make sure i understood it correctly. thanks for your quick response.

@timlacroix
Copy link
Contributor

So, this would only wikidata results, since this eval is not used for ICEWS.
Evaluating on wikidata would take more time. I will try to see if the results change drastically.
It is very possible that there is very little to filter on these datasets anyway so this fix wouldn't change the results significantly.

@timlacroix
Copy link
Contributor

yeah, sorry for the confusion in my answers, it's been a while since I've looked at the code. Thanks for spotting the mistake !

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants