分类目训练商品 word2vec。
logger.info("Training...")
logger.info(" history_index len :%s" % len(self.history_index))
logger.info(" start Word2Vec ... ")
model = Word2Vec(self.history_index, size=representation_size, window=window_size, min_count=4, sg=1, hs=1,
workers=multiprocessing.cpu_count(), iter=1)
logger.info(" got model ...")
model.wv.save_word2vec_format(embedding_file)
logger.info('Train end')
word2vec 开始了几十分钟,还没有结束
2020-07-06 15:58:20,428 - PID:24879 - item_embedding.py[line:26] - INFO: loaded 464 rows from Hive
2020-07-06 15:58:20,429 - PID:24879 - item_embedding.py[line:50] - INFO: Training...
2020-07-06 15:58:20,430 - PID:24879 - item_embedding.py[line:51] - INFO: history_index len :464
2020-07-06 15:58:20,430 - PID:24879 - item_embedding.py[line:53] - INFO: start Word2Vec ...
日志一直停在这里,后来看了一下样本,发现只有一个pid。