Word Embedings

(Gerardo Garcia) #1

@rachel On the presentation of Word Embeddings

After you load the numpy files

vecs = np.load(“glove_vectors_100d.npy”)
vecs50 = np.load(“glove_vectors_50d.npy”)

There’s a section that you load two files

with open(‘words.txt’) as f:
content = f.readlines()
words = [x.strip() for x in content]


wordidx = json.load(open(‘wordsidx.txt’))

Where are those files coming from?
Where is the interface to generate those files?

only includes a text files that looks