reader.py 文件源码

def get_pretrained_embeddings(filepath):
  embeddings = dict()
  with open(filepath, 'r') as f:
    for line in f:
      info = line.strip().split()
      #lines = [line.strip().split() for line in f.readlines()]
      #embeddings = dict([(line[0], [float(r) for r in line[1:]]) for line in lines])
      embeddings[info[0]] = [float(r) for r in info[1:]]
    f.close()
  embedding_size = len(embeddings.values()[0])
  print 'Embedding size={}'.format(embedding_size)
  embeddings[START_MARKER] = [random.gauss(0, 0.01) for _ in range(embedding_size)]
  embeddings[END_MARKER] = [random.gauss(0, 0.01) for _ in range(embedding_size)]
  if not UNKNOWN_TOKEN in embeddings:
    embeddings[UNKNOWN_TOKEN] = [random.gauss(0, 0.01) for _ in range(embedding_size)]
  return embeddings