clips / dutchembeddings Goto Github PK

There is no acces to the word embeddings with the links provided.
They result in the following webpage:
https://www.clips.uantwerpen.be/dutchembeddings/combined-320.tar.gz
which shows:

Forbidden
You don't have permission to access this resource.

Apache/2.4.18 (Ubuntu) OpenSSL/1.0.2g mod_wsgi/4.3.0 Python/2.7.12 mod_perl/2.0.9 Perl/v5.22.1 Server at www.clips.uantwerpen.be Port 443

I'm quite desperate, I planned on using these sets in my bachalor thesis of my study (artificial intelligence, Utrecht University), but now I just found out that I don't have acces. Is there a way to receive this acces or the datasets?

Pascal Verkade
Bachelor student AI @ Utrecht University

Hi, thanks for sharing these materials. For me the usage example with gensim did not work and I found out it is because in newer versions of gensim the load_word2vec_format() function is moved under gensim.models.KeyedVectors instead of under Word2Vec.

Hello, I was wondering if the token 'unk' in the COW models corresponds to the vector trained for unknown tokens. Or is it just a word encountered in the corpus? Best regards.

The download links for all the embeddings are not working.

Hoi,

Ik heb een probleem met het gebruiken van de embeddings. Ik weet haast zeker dat het met mijn lokale computer te maken heeft. Ik denk dat het om de overcomitting ratio gaat. Nou weet ik niet hoe of dat ik deze kan aanpassen in windows.

Kan een van jullie mij misschien op weg helpen?

Groeten,

Giovanni