I did the following changes during my using this lib, but i'm not sure if bugs involved in my changes.
OS: Mac
Python: 3.8
steps:
install
changed requirements.txt from
psutil>=2.1.1
networkx>=2.0
into
psutil>=5.8.0
networkx>=2.6.3
to fix the dependencies analysis issue
init env
remove the line from gensim.models.word2vec import Vocab from skipgram.py (line 8)
as the Vocab doesn't belong to the gensim.models.vord2vec any more, and it will cause a compile issue
run the function
migrate
model = Word2Vec(walks, size=args.representation_size, window=args.window_size, min_count=0, sg=1, hs=1, workers=args.workers)
into
model = Word2Vec(walks, vector_size=args.representation_size, window=args.window_size, min_count=0, sg=1, hs=1, workers=args.workers)
as the param of size doesn't in word2vec.py, and I'm not sure if the representation_size is the vector_size as the description of api: vector_size : int, optional, Dimensionality of the word vectors.
I did the following changes during my using this lib, but i'm not sure if bugs involved in my changes.
OS: Mac Python: 3.8
steps:
into
to fix the dependencies analysis issue
init env remove the line
from gensim.models.word2vec import Vocab
fromskipgram.py
(line 8) as theVocab
doesn't belong to thegensim.models.vord2vec
any more, and it will cause a compile issuerun the function migrate
model = Word2Vec(walks, size=args.representation_size, window=args.window_size, min_count=0, sg=1, hs=1, workers=args.workers)
intomodel = Word2Vec(walks, vector_size=args.representation_size, window=args.window_size, min_count=0, sg=1, hs=1, workers=args.workers)
as the param of size doesn't in
word2vec.py
, and I'm not sure if therepresentation_size
is thevector_size
as the description of api: vector_size : int, optional, Dimensionality of the word vectors.