You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on May 30, 2023. It is now read-only.
However it doesn't seem to work - cliff does not recognize even one-word-requests anymore (like q=Krems, a small town in Austria). The same requests would be resolved with language=EN.
I am running on a local machine with 16GB RAM and from the cliff-docker.
Is it just that the ner in German is much worse? Are there any known workarounds?
Is it a RAM issue?
The text was updated successfully, but these errors were encountered:
I've done only initial testing in German, which I don't speak, so I can't characterize how well it works. That said, I just tried a short sentence from a Der Spiegel article on our homepage and found that it pulled out a place and person pretty well - results below. You are correct that the one-word request for Krems doesn't work. I expect that is because the models are built to process sentences, but I'm not sure 🤷🏽♂️.
Input (from this article)
"Alesja* sieht noch, wie Beamte einer Omon-Sondereinheit mit Schlagstöcken auf ihren Mann einprügeln. Iwan* liegt mit anderen auf der Straße an einer Brücke in Minsk, hält die Hände über den Kopf verschränkt. "
According to https://cliff.mediacloud.org/, cliff can also geoparse German text. Assuming that I can tell the API via ?language=DE from the form params in the calls on https://cliff.mediacloud.org/.
However it doesn't seem to work - cliff does not recognize even one-word-requests anymore (like
q=Krems
, a small town in Austria). The same requests would be resolved with language=EN.I am running on a local machine with 16GB RAM and from the cliff-docker.
The text was updated successfully, but these errors were encountered: