訓練 word2vec 詞向量時，內存不足如何解決？

最近在學習自己訓練詞向量,整理了13GB的小說語料，昨晚設置好從分詞做到詞向量訓練，結果今天早上看了一下內存溢出。僅僅只完成了130M的分詞，連詞向量的訓練都沒開始就宕機了。請問是我代碼有問題還是正常現象？　下圖為執行代碼分詞工具使用的是北大的pkuseg如果是代碼問題請問如何應如何修改？
sentences = []
rules = u"[u4e00-u9fa5]+"

pattern = re.compile(rules)
f_writer = open("data/分詞後的1s1lCN13G.txt", "w", encoding="utf-8")
with open("data/1s1lCN13G.txt", "r" , encoding="utf-8") as f_reader:
for line in f_reader:

# line = line.replace("
","").replace("
","").strip()
if line == "" or line is None:
continue
line = " ".join(seg.cut(line))
seg_list = pattern.findall(line)

word_list = []
for word in seg_list:
if word not in stop_words:
word_list.append(word)
if len(word_list) &> 0:

sentences.append(word_list)
line = " ".join(word_list)
f_writer.write(line + "
")
f_writer.flush()

f_writer.close()
#模型訓練
model = word2vec.Word2Vec(sentences,sg=1,hs=1,window=10,min_count=20,iter=10,size=256,workers=16)
model.save("data/13G")