def loadX(fname):
'''
Read data records into a data matrix.
Also return vocabulary.
'''
events = []
words_keys = set()
for e in load(fname):
events.append(e)
words_keys = words_keys | set(e[5].keys())
words_keys = sorted(list(words_keys))
for (eidx, e) in enumerate(events):
events[eidx] = list(e[:5]) + [e[5].get(word_key, 0)
for word_key in words_keys]
X = np.vstack(events)
return (X, words_keys)
评论列表
文章目录