将在python2下生成的pkl文件转换成可被python3读取的pkl

import os
import dill
import pickle
old_pkl_file = " "
new_pkl_file = " "
dill._dill._reverse_typemap["ObjectType"] = object

with open(old_pkl_file, 'rb') as f:
    loaded = pickle.load(f, encoding="latin1")

with open(new_pkl_file, 'wb') as outfile:
    pickle.dump(loaded, outfile)

(1) pickle文件为二进制文件，因此读写的时候需要'rb'或'wb'。

(2) 在python2转到python3的数据会报编码错误，例如：“UnicodeDecodeError: 'ascii' codec can't decode byte 0xb1 in position 8: ordinal not in range(128)” 因此需要在load函数上加上编码：encoding="latin1"。

贴上python3官网上关于此的描述如下：链接