The _raw_word_count method is breaking word2vec model

[edilsonacjr]

The old code for the function _raw_word_count is breaking the word2vec model:

File "/Library/Frameworks/Python.framework/Versions/3.5/lib/python3.5/site-packages/gensim/models/word2vec.py", line 744, in
return sum(len(sentence) for sentence in job)
TypeError: object of type 'map' has no len()

The solution was already implemented. Why is it not yet available?

From (Issue #535 ):
def _raw_word_count(self, job):
"""Return the number of words in a given job."""
return sum(len(sentence.words) for sentence in job)

[tmylk]

Dear @edilsonacjr

What is the type of input that you are giving to sentence? Really keen to reproduce the error. Do you have a code snippet?

The code is current for last 12 months. In (word2vec)[https://github.com/RaRe-Technologies/gensim/blame/2a70e3a726404cd4230542a35cfd2dc4d63da6f1/gensim/models/word2vec.py#L747] len(sentence) was added in #535. The change to len(sentence.words) only affects doc2vec and not word2vec.

[edilsonacjr]

I just realized the problem is that I'm using the 'map' function directly as sentences, instead of a list:

walks = [map(str, walk) for walk in walks]
model = Word2Vec(walks, size=args.dimensions, window=args.window_size, min_count=0, sg=1, workers=args.workers,
                 iter=args.iter)
model.save_word2vec_format(args.output)

Thank you anyway.

[tmylk]

Glad you figured it out! Closing as "not an issue".