Why not cosine similarity?

[mfxss]

In explore_context2vec.py, when got context_v and target_v, why use w again?
You use the w to get the context_v, didn't you?

def mult_sim(w, target_v, context_v):
    target_similarity = w.dot(target_v)
    target_similarity[target_similarity<0] = 0.0
    context_similarity = w.dot(context_v)
    context_similarity[context_similarity<0] = 0.0
    return (target_similarity * context_similarity)

[orenmel]

target_v is the vector representation of the target word
context_v is the vector representation of the context
w is a matrix that comprises the vectors of all words in the vocabulary. In mult_sim it is used to compute the similarities between target_v and all words (target_similarity), and then between context_v and all words (context_similarity). Finally, the product of target_similarity and context_similarity is used to rank all the words in w.