search

VHRanger / nodevectors

Fastest network node embeddings in the west
MIT License
506 stars 59 forks source link

word2vec parameters changed #49

Closed daniel-unyi-42 closed 2 years ago

daniel-unyi-42 commented 2 years ago

Hi!

In node2vec.py, you should modify the 'iter' parameter to 'epochs' and the 'size' parameter to 'vector_size'.

(And thank you for the library, I use it extensively in my research!)

VHRanger commented 2 years ago

The parameters in the Node2Vec class are to stay consistent with SKLearn conventions (n_components and epochs)

Do you mean the ones passed to gensim are erroneous?

daniel-unyi-42 commented 2 years ago

Yes, the parameter names in the constructor of gensim.models.Word2Vec were changed. Sorry for the misunderstanding.

VHRanger commented 2 years ago

Ah, actually it's tracked here:

https://github.com/VHRanger/nodevectors/issues/37

I'll close this one and resolve the other one right away