#!/usr/bin/env python3

import gzip
import numpy
import pickle

vectors = {}
with gzip.open("numberbatch-en-19.08.txt.gz", mode="rt") as f:
    count = int(next(f).split()[0])
    for (i, line) in enumerate(f):
        if i % 1000 == 0:
            print("{}/{}".format(i, count))
        [word, *vec] = line.split()
        vectors[word] = numpy.array([float(x) for x in vec])

with open('numberbatch.pkl', 'wb') as f:
    pickle.dump(vectors, f)